Clinician Attitudes and Beliefs About Deintensifying Head and Neck Cancer Surveillance

This qualitative study surveys otolaryngologists and radiation oncologists regarding their opinions on deintensification of surveillance in patients with diagnosed head and neck cancer.

Key Points

Question

What are otolaryngologists’ and radiation oncologists’ perspectives on deintensifying head and neck cancer surveillance?

Findings

In this qualitative study of 21 otolaryngologists and radiation oncologists, clinicians identified potential barriers to deintensifying surveillance, including patient and physician peace of mind, need to maintain a physician-patient relationship, and need to adequately manage treatment-associated toxic effects and other survivorship concerns.

Meaning

These findings suggest that incorporation of surveillance and survivorship education in training, positive reframing of surveillance deintensification, and creation of virtual survivorship programs may address some barriers to deintensification.

Abstract

Importance

Surveillance imaging and visits are costly and have not been shown to improve oncologic outcomes for patients with head and neck cancer (HNC). However, the benefit of surveillance visits may extend beyond recurrence detection. To better understand surveillance and potentially develop protocols to tailor current surveillance paradigms, it is important to elicit the perspectives of the clinicians who care for patients with HNC.

Objective

To characterize current surveillance practices and explore clinician attitudes and beliefs on deintensifying surveillance for patients with HNC.

Design, Setting, and Participants

This qualitative study was performed from January to March 2021. Guided by an interpretive description approach, interviews were analyzed to produce a thematic description. Data analysis was performed from March to April 2021. Otolaryngologists and radiation oncologists were recruited using purposive and snowball sampling strategies.

Main Outcomes and Measures

The main outcomes were current practice, attitudes, and beliefs about deintensifying surveillance and survivorship as well as patients’ values and perspectives collected from interviews of participating physicians.

Results

Twenty-one physicians (17 [81%] men) were interviewed, including 13 otolaryngologists and 8 radiation oncologists with a median of 8 years (IQR, 5-20 years) in practice. Twelve participants (57%) stated their practice comprised more than 75% of patients with HNC. Participants expressed that there was substantial variation in the interpretation of the surveillance guidelines. Participants were open to the potential for deintensification of surveillance or incorporating symptom-based surveillance protocols but had concerns that deintensification may increase patient anxiety and shift some of the burden of recurrence monitoring to patients. Patient and physician peace of mind, the importance of maintaining the patient-physician relationship, and the need for adequate survivorship and management of treatment-associated toxic effects were reported to be important barriers to deintensifying surveillance.

Conclusions and Relevance

In this qualitative study, clinicians revealed a willingness to consider altering cancer surveillance but expressed a need to maintain patient and clinician peace of mind, maintain the patient-clinician relationship, and ensure adequate monitoring of treatment-associated toxic effects and other survivorship concerns. These findings may be useful in future research on the management of posttreatment surveillance.

Introduction

Improvements in overall survival and the increasing incidence of human papilloma virus (HPV)–positive oropharyngeal cancer have resulted in an increase in the population of survivors of head and neck cancer (HNC) in the US.¹ The National Comprehensive Cancer Network guidelines provide broad recommendations for HNC that do not risk stratify patients by clinical characteristics, such as HPV infection status or site.² The guidelines recommend surveillance visits every 1 to 3 months during the first year, 2 to 6 months during the second year, 4 to 8 months during years 3 to 5, and annually thereafter. Posttreatment imaging is recommended within 6 months of treatment for patients with locoregionally advanced disease. Although the guidelines provide a broad overview, current surveillance practices have been shown to be highly variable among clinicians.^3,4,5

Studies have demonstrated that routine visits and imaging are costly and do not provide a survival advantage for patients with HNC.^{3,6,7,8,9,10,11,12,13} Furthermore, most recurrences are diagnosed based on symptoms rather than on examination findings for asymptomatic patients.^7,9,10,11 Patient surveys have suggested that many would prefer decreasing the frequency of surveillance visits,¹⁴ and those who experience financial toxicity are less likely to be adherent to recommended follow-up and treatment.¹⁵ In light of this information, studies have suggested risk-stratifying surveillance and deintensifying regimens for patients with low risk of recurrence, particularly those with HPV-associated disease.^10,11 However, the value of surveillance visits to clinicians may extend beyond recurrence detection and include survivorship care as well as enhancement of the patient-physician relationship.

To better understand HNC surveillance and potentially develop protocols to tailor the current surveillance paradigm, it is important to understand the perspectives of the clinicians who provide care for patients with HNC with respect to surveillance. The purpose of our study was to characterize current surveillance practices and explore clinician attitudes and beliefs on deintensifying surveillance.

Methods

For this qualitative study, we conducted semistructured interviews with clinicians whose specialty was HNC throughout the US from January to March 2021 through video conferencing (Zoom; Zoom Videocommunications). Eligible participants were otolaryngologists or radiation oncologists who treated patients with HNC. Participants were recruited via email from a personal and professional network. Purposive sampling was used to focus on clinicians with varying experience levels, proportions of patients with HNC in their practices, case volumes, and practice types. We then used snowball sampling by asking each participant for referrals of other clinicians who treated patients with HNC to recruit additional participants. The study was determined to be exempt from review by the University of Michigan institutional review board because it involved only interview procedures and the information recorded was deidentified. All participants verbally consented to be interviewed and were offered a $25 gift card as an incentive. This study followed the Standards for Reporting Qualitative Research (SRQR) guideline.

The semistructured interview guide (eAppendix in the Supplement) was developed iteratively in collaboration with subject and methodologic experts. We conducted 1 pilot interview with a clinician who met study eligibility criteria and whose feedback informed slight modifications to the subsequent interview guide. The pilot interview was not included in the results. In brief, participants were asked about their current practice, attitudes and beliefs about deintensifying surveillance, views on survivorship, and patients’ values and perspectives. In addition, a demographic survey was administered to obtain information on years of practice, volume of HNC cases, and new patient volume.

Two interviewers (M.M.C. and N.M.M.) trained in qualitative methods conducted all interviews. Interviews were audiorecorded, transcribed verbatim, and deidentified. We followed the inductive and iterative approach of interpretive description, a qualitative method that interprets participants’ subjective experiences to improve understanding of clinical issues.¹⁶ We used information power to estimate and assess the sample size.¹⁷ This assessment was guided by the fact that the study aim was narrow, the sample was dense, and the quality of the dialogue was strong and clear. Transcripts were imported to MAXQDA, version 2020 software (VERBI Software) to support coding and analysis. The research team developed a codebook that contained structural and descriptive codes deductively applied for each question. These codes were later supplemented by inductively derived codes descriptive of factors volunteered by the participants. Each interview was coded independently by 2 researchers (M.M.C. and N.M.M.). The coding was discussed at weekly team meetings, and any differences in the researchers’ coding were resolved through discussion until a consensus was met. Constant comparison was used to refine the coding taxonomy.¹⁸

We used data abstraction, case comparison, and memo writing to focus and develop our themes. We retrospectively applied the Theoretical Domains Framework to systematically group our themes into domains.¹⁹ The Theoretical Domains Framework is an integrated theoretical framework that is used in health behavior change and implementation research. These domains were also mapped to the Capability, Opportunity, Motivation, and Behavior Model components.²⁰ This model is used to evaluate how components of capability, opportunity, and motivation interact with behavior and can be used to identify potential targets for behavioral interventions.²⁰ To address trustworthiness, we also discussed alternative interpretations, biases, latent themes, prevalence, outliers, and the clinical implications of our findings in regular meetings. Data analysis was performed from March to April 2021.

Results

Study Participants

Twenty-one physicians were interviewed, including 13 otolaryngologists and 8 radiation oncologists (Table 1). Of these, 17 were men (81%) and 4 were women (19%). The median time in practice was 8 years (IQR, 5-20 years); 11 physicians (52%) had less than 10 years of experience, and 10 physicians (48%) had 10 or more years of experience. Most clinicians (19 [90%]) practiced in an academic setting, and 12 individuals (57%) stated that patients with HNC comprised more than 75% of their practice.

Table 1. Participant Demographic Characteristics.

Characteristic	Participants, No. (%)
Sex
Female	4 (19)
Male	17 (81)
Specialty
Otolaryngology	13 (62)
Radiation oncology	8 (38)
Region
East	3 (14)
Midwest	14 (67)
West	3 (14)
South	2 (10)
Experience
Median (IQR), y	8 (5-20)
<10 y in practice	11 (52)
≥10 y in practice	10 (48)
Practice setting
Academic	19 (90)
Community	2 (10)
Head and neck cancer case volumea
≤50%	4 (19)
51%-75%	5 (24)
>75%	12 (57)
New patient volumeb
<100	5 (24)
100-150	8 (38)
>150	7 (33)
Unknown	1 (5)

^a Defined as annual percentage of patients with head and neck cancer.

^b Defined as annual number of new patients with head and neck cancer.

Current Surveillance Practice

Clinicians reported a varied frequency of surveillance visits, ranging from every 6 weeks to every 4 months for the first 2 years and then every 3 to 6 months until 5 years after treatment. After 5 years, some clinicians evaluated patients annually, and others recommended them to be seen only as needed. Clinicians would often alternate visits between otolaryngology (head and neck surgical oncology trained) and radiation oncology but were more reluctant to alternate surveillance visits with local general otolaryngologists and primary care physicians. Many performed endoscopic examinations at each surveillance visit; the remainder varied based on the clinical scenario (eg, disease site, stage of tumor, and patient risk factors) and increased the interval between endoscopic examinations over time or performed mirror examinations. Clinicians also discussed performing only mirror examinations and not routinely performing endoscopic examinations for patients with oral cavity sites unless the patient smoked, was symptomatic, or had a difficult mirror examination.

Most participants obtained a 3-month posttreatment computed tomography neck and chest scan or a positron emission tomography scan and no further routine imaging unless prompted by symptoms or clinical concern. Many participants discussed previously having obtained routine annual imaging but did not find benefit and had since limited routine scanning to 1 posttreatment scan. The remainder of clinicians obtained images routinely every 6 to 12 months. Many clinicians also reported routinely obtaining annual low-dose chest computed tomography scans for all patients who smoked.

Guidelines

All participants were aware of the National Comprehensive Cancer Network guidelines but reported that the broad nature of the guidelines often led to variation in their practice. Many approached the guidelines as a suggested practice but discussed perceived room for interpretation based on clinical judgment. Several clinicians discussed a lack of substantial evidence supporting the guidelines with respect to surveillance visits, imaging, and endoscopic examinations. Clinicians stated that their current or past institutional culture often influenced their practice patterns.

Physician-Level Factors

Participants expressed mixed feelings about deintensifying cancer surveillance. In general, clinicians with 10 or more years of experience had favorable attitudes and beliefs about deintensifying surveillance, stating that in their experience, recurrences were commonly associated with symptoms and a more flexible surveillance schedule was more patient-centered. Participants were also more comfortable with symptom-based surveillance after the first 2 to 3 years for select patients, such as those with HPV-positive oropharyngeal cancer and older individuals or those with substantial financial or travel burdens and low-risk disease. Clinicians spoke of their personal clinical experience and the rarity of detecting a recurrence during physical examination of an asymptomatic patient. However, participants acknowledged that standardization across all patients and having a more consistent algorithm was easier to incorporate into clinical workflows. A few participants discussed a preference to have more data guiding their decision-making to risk-stratify patients (Table 2).

Table 2. Theoretical Domains Framework for Behavior Change Mapped to Clinician-Level Factors With Exemplary Quotes^a.

Domain and theme	Sample quote (participant identifier, clinician specialty)
Knowledge
Clinical experience leads to more comfort with deintensification	“I feel strongly about it, you know, still seeing them frequently, and maybe that's because I'm more junior and quite afraid of recurrences.” (P7, radiation oncology) “I'm, you know, 4 years into my career now; and so I think, maybe, as I probably get a little bit older I'll probably be able to cut back too a little bit.” (P8, radiation oncology)
Clinician training and inertia in practice patterns	“We're all victims of our training.” (P3, radiation oncology) “It's hard to change your customs; the ones that you've been trained at.” (P20, radiation oncology)
Skills
Endoscopic examinations are a specialized skill	“It's not the same exam in every clinic and then with every colleague...not everybody, especially in the radiation oncology world, is comfortable performing flexible laryngoscopy. So if an exam warrants a flexible laryngoscopy, a lot of our radiation oncology colleagues and the medical oncologists as well will send the patient back to us.” (P3, otolaryngology) “Head and neck cancer is too esoteric of a disease for a primary care provider to manage.” (P11, otolaryngology)
Clinicians’ discomfort addressing survivorship	“As oncologists we're good at managing like immediate direct psychosocial harms and physical harms to patients from tumor and treatment, but we're not good at managing a lot of the other stuff that is associated with either being elderly in general or having had this cancer.” (P12, otolaryngology) “Many of us aren't necessarily skilled to provide all the different aspects that go into mental, emotional health.” (P19, otolaryngology)
Decision
Standardization is easier	“I do think there's some value in having standardization, which is why I try not to veer from it too much.” (P6, otolaryngology)
Social and professional role and identity
Uncertainty in who is managing survivorship	“I think ultimately that the head and neck surgeon should, in a surgical patient, take the lead [in surveillance/survivorship]. In a nonsurgical patient, you know, perhaps the med[ical] onc[ologist] or the rad[iation] onc[ologist].” (P11, otolaryngology)
Beliefs about capabilities
Trust in local otolaryngologists	“I don't totally trust the outside docs as much.” (P1, otolaryngology) “Yeah, I would say the opposite. I think most of the time, I have a great deal of trust in the local providers...these are trained, well trained usually. They're board-certified individuals and many of them are doing a lot of care...so I think [mistrust] seems to be somewhat misplaced.” (P9, otolaryngology)
Reinforcement
Importance of financial incentives	“Today if we told otolaryngologists we're not gonna let you do a flex scope on some head and neck cancer patients more than twice a year, a lot of head and neck cancer patients would start to be seen twice a year.” (P9, otolaryngology) “There are absolutely scenarios where our practices are driven in part due to reimbursement, and I think that's a shame. But I don't think this is one of them.” (P5, radiation oncology)
Fear of litigation	“Certainly litigation is never far from the front of people's minds.” (P21, otolaryngology)
Emotion
Clinician anxiety and need for peace of mind	“I get more and more cautious about decreasing the surveillance...it's just such a low-hanging fruit to catch a recurrence and is something that can make such a huge difference.” (P7, radiation oncology) “I think a deeper reason is fear of, you know, inadequacy. So they're not comfortable with, you know, the exam. They're not comfortable interpreting the symptomatology.” (P15, otolaryngology)
Preservation of patient-physician relationship	“So much of my job is my relationship with my patients. So I think that surveillance, while it has its role on a practical level of trying to find cancer, it has this other intangible role of like what you get out of your job as a physician that you don't even realize that you value about your job, which is basically saying 'hi' to this face that you've saved their life.” (P18, otolaryngology) “I went into head and neck because I like the surgeries, but also because I like the patient population. I like the relationships with the patients and the longevity.” (P14, otolaryngology)

^a The Theoretical Domains Framework is an integrated theoretical framework that is used in health behavior change and implementation research.¹⁹

Clinicians with less than 10 years of experience were less supportive of altering surveillance. They specifically cited factors including concern about missing recurrences or not detecting recurrences early enough and the need for frequent visits for close monitoring of treatment-associated toxic effects. Some participants were more comfortable with considering altering surveillance 2 or more years after treatment. Participants stated that some of this discomfort may be owing to being both risk averse and early in their career. They also stated that their practice reflected their experience in training and discussed habit or a sense of inertia as barriers to changing their current practice. Concerns about symptom-based surveillance centered on the need for more data on using quality-of-life questionnaires and telemedicine as the primary metrics for aiding symptom-based surveillance and detecting concern for recurrence. In addition, clinicians expressed worry that symptom-based surveillance may increase patient anxiety and require more patient education and reassurance; in addition, they were concerned about patient reliability in reporting symptoms and missing recurrences.

Clinicians reported that maintaining a patient-physician relationship was one of the most important and fulfilling aspects of providing care for patients with HNC. Many discussed a psychosocial advantage of maintaining that relationship through surveillance visits even if they did not alter the rate of detection of recurrences. Clinicians highly valued the long-term relationship and trust with their patients and did not want to give up that aspect of their practice.

The value of alternating with local general otolaryngologists or radiation oncologists was commonly acknowledged as a benefit especially for patients with financial or travel concerns. However, factors that influenced willingness to alternate with local clinicians included their trust and relationship with the clinician regardless of whether the local clinician performed endoscopic or mirror examinations and a desire to follow up with patients in their care regularly to have context for changes in examination findings. Clinicians rarely alternated visits with local medical oncologists. Clinicians were more willing to alternate visits with local primary care physicians 5 or more years after treatment when the focus was more on survivorship than on surveillance.

Patient-Level Factors

Most clinicians suggested that specific patient groups with lower risk of recurrence, such as those with HPV-associated disease, could benefit from deintensified surveillance. Some also suggested that deintensification would be most helpful for patients at low risk who found frequent visits to be a hardship, such as older patients, those with a substantial travel burden, or those with limited finances or resources (Table 3). Patient reliability in reporting new symptoms was also an important factor to clinicians when thinking about deintensifying surveillance. Clinicians expressed concern about deintensifying surveillance for individuals who smoked, those without HPV infection, and patients with HNC sites in which symptoms of recurrence are less likely to arise early, such as the hypopharynx.

Table 3. Theoretical Domains Framework for Behavior Change Mapped to Patient- and System-Level Factors With Exemplary Quotes^a.

Domain and theme	Sample quote (participant identifier, clinician specialty)
Patient level
Social and professional role and identity
Patient expectations for their clinicians	“Patients don't want to lose that contact with their primary oncology person either and don't want to just be shunted off to, you know, someone that they don't know.” (P15, otolaryngology)
Behavioral regulation
Patient education modulates expectations	“I think the level of patient concern may be related to how it's discussed with them.” (P2, radiation oncology) “I think there's a small group of people who have a very specific idea [about surveillance], but I think by and large we as providers inform what that is.” (P19, otolaryngology)
Beliefs about capabilities
Patient reliability concerns	“I don't trust my patients. Some of them are very anxious. But there's a lot of them that I'm like, did you not notice that there's like a golf ball inside of your mouth? Like that has probably not been there for only a week.” (P14, otolaryngology)
Social influences
Patient travel and financial concerns	“I think [travel and financial concerns] are really natural barriers and probably very underappreciated barriers. It is real and people are embarrassed. They don't necessarily want to relay that to you. I think it's an important thing to be aware of when somebody says, ‘Hey, can I skip a visit?’ [and delve] into why.” (P19, otolaryngology)
Lack of patient support system	“The patient population I deal with is a little rough around the edges, should we say, like many head and neck patients. I do not have a population of patients who have family support and resources and things. A lot of them refuse to see primary care. They don't have anybody else who's looking after them, except for us.” (P14, otolaryngology)
Emotion
Patient anxiety and need for peace of mind	“I've sent a few HPV-positive, nonsmokers to the front desk and been like, 'You don't have to come back for 6 months.' Then they appear again on my schedule in 2 or 3 months. So, patient anxiety, while it may not directly play into things, it indirectly comes into surveillance.” (P2, radiation oncology) “Sometimes they want to be seen because they're scared.” (P5, radiation oncology)
Preservation of the patient-physician relationship	“I think one thing that terrifies all patients is to be abandoned by their providers. We take care of head and neck cancer patients at their most vulnerable and promise them that we're going to follow them and get them through their treatment and indefinitely thereafter.” (P4, otolaryngology)
System level
Knowledge
Lack of evidence supporting current guidelines	“I personally feel a little trapped by guidelines, like the NCCN guidelines that are simply guidelines and around follow-up. I would challenge the people that write those to show me 1 good paper that really suggests that all these visits make any difference at all.” (P9, otolaryngology) “When we look at the guidelines, you know, they leave a lot for interpretation, you know, 3-6 mo?” (P10, radiation oncology) “I don't know and I'm not aware of any evidence that our current surveillance regimen, you know, that sort of NCCN dictated actually has resulted in any improvement in survival in any patients.” (P12, otolaryngology)
Lack of evidence supporting current practice	Visits: “I'm not aware of there being any survival benefit to the current regimen that we're employing.” (P12, otolaryngology) “I think, if anything, I probably oversee patients honestly. But the reality is we don't really have a good answer to that.” (P17, otolaryngology) Scopes: “I'm not even sure an annual scope tells us anything.” (P3, radiation oncology) Imaging: “I used to do a 1-y PET scan, never found anything, never found it was useful, and so I've scaled back most of my surveillance imaging.” (P1, otolaryngology)
Environmental context and resources
Surveillance and survivorship are linked	“We are over-surveilling these patients, but probably under-survivorshipping them.” (P2, radiation oncology) “Thinking about surveillance as only detecting second primaries and recurrences is shortsighted...surveillance and survivorship care are much broader than that. So we need to make sure that we're treating not just the cancer, but the patient.” (P4, otolaryngology) “Cancer recurrence is obvious. I think that evaluating toxicity and quality of life is the more contentious, the more complex thing.” (P16, radiation oncology)
Telemedicine has a limited role	“I don't prefer a phone call or a video visit to seeing someone in person where I can examine them.” (P13, otolaryngology) “Video visits are not so ideal for head and neck cancer surveillance.” (P8, radiation oncology)

Abbreviations: HPV, human papilloma virus; NCCN, National Comprehensive Cancer Network; PET, positron emission tomography.

^a The Theoretical Domains Framework is an integrated theoretical framework that is used in health behavior change and implementation research.¹⁹

Patients’ need for peace of mind was described as a barrier to deintensifying surveillance. Clinicians described scheduling more frequent visits or ordering scans even when not clinically indicated to address patient anxiety. With the current gradual lengthening of surveillance intervals, participants believed patients still had substantial concerns about decreasing the visit frequency owing to fear of recurrence. Participants stated that more symptom-driven surveillance may increase patient anxiety and may transfer to patients the psychological burden of monitoring their symptoms. In general, clinicians stated that patients expect physicians to provide guidance on surveillance expectations. Some reported that patient education on risk of recurrence and setting expectations about surveillance may reassure patients and alleviate patient anxiety about altered surveillance.

System-Level Factors

Most clinicians had used some form of telemedicine as a result of the COVID-19 pandemic. Clinicians reported using telemedicine as a screening visit or triage tool, a way to alternate with in-person visits, for symptom management and survivorship, in conjunction with in-person visits with a local clinician, or for symptom follow-up for patients receiving palliative care. Some clinicians did not think they would continue to incorporate telemedicine heavily in their surveillance practice after the pandemic owing to the inability to perform physical and endoscopic examinations in the virtual format, potential lack of continuing reimbursement, or uncertainty of how telemedicine fits in their typical practice. Participants noted that when offered virtual visits, many patients preferred in-person visits.

Clinicians acknowledged that surveillance and survivorship have different goals but that in the current paradigm, surveillance visits and intervals were often used to address both surveillance for disease recurrence and management of treatment-associated toxic effects and other survivorship concerns. Owing to the linkage of survivorship and surveillance, clinicians suggested that the current practice may lead to unnecessary surveillance (“over-surveilling”) but a lack of attention to survivorship concerns (“under-survivorshipping”) of patients with HNC. Many clinicians discussed feeling they did not have the tools or time to adequately manage the wide range of survivorship issues, particularly psychosocial issues such as depression, social intimacy, and body image concerns.

Discussion

This qualitative study explored clinicians’ current surveillance practices for patients with HNC and their attitudes and beliefs about tailoring surveillance strategies in the future. Participants expressed that there was much room for interpretation of the current guidelines. Clinicians were open to the potential for deintensifying surveillance or incorporating symptom-based surveillance but had concerns that deintensifying may increase anxiety and shift some of the burden of recurrence monitoring to patients. Peace of mind, maintenance of the patient-physician relationship, and the need for adequate survivorship and management of treatment-associated toxic effects were important barriers to altering surveillance (Box).

Box. Barriers and Facilitators for Deintensifying Head and Neck Cancer Surveillance.

Barriers

Clinician level

• Less than 10 years in practice

• Importance of examination and flexible laryngoscopy

• Desire for standardized protocol

• Desire to be actively involved in a patient's cancer care and variable trust in local clinicians

• Need for peace of mind and concern for recurrence and second primary tumors

• Maintenance of patient-physician relationship

• Financial incentives

• Fear of litigation

Patient level

• Expectation for routine surveillance

• Need for peace of mind and anxiety about recurrence

• Reliability, travel and financial concerns, and lack of support network

• High-risk disease (eg, advanced-stage disease, disease in persons who smoke tobacco)

• Maintenance of patient-physician relationship

System level

• Need for more research

• Limited use of telemedicine for surveillance examinations

• Surveillance visits linked to survivorship visits

Facilitators

Clinician level

• Ten years or more in practice

• Recurrences less likely to be noted in examinations of asymptomatic patients

• Recognition of different risk factors and needs of patients

• Established relationships with local clinicians and primary care physicians

Patient level

• Patient education about surveillance needs

• Patient education about risk of recurrence

• Ancillary services and a strong support network

• Low-risk disease (eg, low-stage disease, human papilloma virus–associated disease, and disease in persons who do not smoke tobacco)

System level

• Lack of evidence supporting current guidelines and practice

• Increased use of telemedicine for symptom screening

Despite awareness of the guidelines, nearly half of the participants obtained routine surveillance imaging and expressed that the recommendations about surveillance visits were broad and left room for variability. Roman et al⁵ reported that, even among clinicians who were aware of the surveillance imaging guidelines, 31% ordered surveillance positron emission tomography scans more than 50% of the time for asymptomatic patients. On multivariable analysis in that study, there was no association between specialty, physician sex, years in practice, practice settings, or HNC case volume and increased use of imaging.

A key emergent theme in our study was the importance of peace of mind in influencing the intensity of cancer surveillance. Patients’ and clinicians’ need for peace of mind have been shown to be associated with overtreatment for other cancers, including prostate,²¹ thyroid,²² breast,²³ and ovarian.²⁴ Physicians in general are trained to be relatively intolerant of uncertainty and error.^25,26 In our study, clinicians with fewer years in practice expressed concerns about deintensification owing to fear of recurrence and also referenced that their practice may change as they advance in their careers. The longer clinicians were practicing, the more likely they were to express favorable attitudes about deintensification because of their personal clinical experience and increased trust with local physicians.

For patients, emotional responses to cancer treatment may lead to choices not necessarily in their best interest.²⁷ Patients’ preferences are sensitive to how options are framed by their physician, and education can help improve tolerance of uncertainty, assuage patient concerns, and provide peace of mind.²⁸ Because most recurrences are identified on the basis of patient symptoms,^7,10,11 positive reframing of symptom-based surveillance may empower patients rather than cause anxiety and may improve patient reliability and early detection. In a survey of patients with low-risk oropharyngeal cancer, 55.2% patients were interested in decreased frequency of in-clinic surveillance visits.¹⁴ This proportion increased to 61.2% after education about the low risk of recurrence.¹⁴ Patient-level factors that were associated with decreased interest in deintensification included being a medical maximizer (individuals who prefer to take an active approach to health care and may want to receive optional medical tests and treatments), having a long-term relationship with the physician, and having higher worry of cancer recurrence.¹⁴

Maintaining the patient-physician relationship was valued by clinicians and was another important barrier to reduced surveillance. The need to maintain this relationship had already resulted in reluctance to alternate surveillance with local otolaryngologists, radiation oncologists, or primary care physicians. Brennan et al²⁹ surveyed 175 patients with HNC and reported that 79.4% preferred to continue to see their oncologist rather than have follow-up with their family physician (85.7%) or exclusively by a clinic nurse (90.3%). However, despite the value clinicians placed in the patient-physician relationship, they acknowledged the travel and financial burden of frequent surveillance for patients and expressed discomfort in addressing the wide range of survivorship issues, particularly psychosocial concerns. This discomfort suggests that, although sympathetic to the patient perspective, clinicians may have viewed the patient-physician relationship through a clinician-centered lens, and a more patient-centered view may allow for more-tailored surveillance schedules.

Potential interventions could be focused on the knowledge, emotion, and environmental context domains (Figure). To address knowledge, more education about surveillance and survivorship could be incorporated into residency and fellowship training so that trainees develop comfort with managing follow-up and have less fear and uncertainty when they start their own clinical practice. Research could investigate the utility of payment incentives for incorporating multidisciplinary survivorship care or patient-centered outcomes in survivorship care. For the emotion domain, positive reframing of deintensified surveillance to patients may help with patient anxiety about deintensification. Communication training may provide clinicians with strategies to actively address patient anxiety and peace of mind other than with additional imaging. Shared decision-making may lead to more patient-centered discussions of surveillance by encouraging team talk of surveillance options with the use of decision aids when possible and may help physicians focus in aiding patients in exploring and forming their personal preferences.³⁰ In addition, in the current environment, surveillance and survivorship provision are linked during routine follow-up visits, resulting in an overuse of surveillance and underuse of survivorship.³¹ Restructuring of this environment by continuing in-person surveillance with surgeons and radiation oncologists and also using telemedicine and virtual care to manage symptoms and create virtual survivorship programs may improve the balance between surveillance and survivorship.

Figure. Proposed Interventions Based on the Capability, Opportunity, Motivation, and Behavior Model and the Theoretical Domains Framework for Behavior Change.

The Theoretical Domains Framework is an integrated theoretical framework that is used in health behavior change and implementation research.¹⁹ The Capability, Opportunity, Motivation, and Behavior Model is used to evaluate how components of capability, opportunity, and motivation interact with behavior and can be used to identify potential targets for behavioral interventions.²⁰

Limitations

This study has limitations. First, selection bias was possible. Most clinicians had primarily head and neck–focused practices and worked at academic medical centers; thus, they may have been more comfortable with deintensifying surveillance. We also did not interview patients; therefore, we did not have data on patient preferences. However, prior work has suggested that patients experience a significant travel and financial burden with surveillance visits, and more than half were interested in decreasing the frequency of surveillance visits.¹⁴ The background and theoretical orientation of the research team may also have affected the results if participants shared only thoughts that they felt the team wanted to hear.

Conclusions

This qualitative study explored clinician perspectives about surveillance of patients with HNC and the potential for deintensification. Clinicians revealed a willingness to consider altering cancer surveillance but expressed a need to maintain patient and clinician peace of mind, maintain the patient-clinician relationship, and ensure that there still was adequate monitoring of treatment-associated toxic effects and other survivorship concerns. These findings might be used to guide future research in the management of posttreatment surveillance and survivorship, specifically through a surveillance deintensification implementation trial and a cost analysis of altering surveillance.

Supplement.

eAppendix. Interview Guide