Using the Delphi Method to Elucidate Patient and Caregiver Experiences of Cancer Care

Introduction

Head and neck cancer (HNC) carries a substantial physical and emotional burden throughout the disease trajectory (1–4). Unmet informational needs and untreated symptoms are associated with increased psychological, existential, and physical distress (5–7). Distress can reduce patients' quality of life and ability to cope with cancer (8), exacerbate symptoms, and correlates with increased Emergency Room visits and hospitalizations (5,9,10) reduced treatment compliance (11), and survival (12,13). Understanding the needs of patients with HNC is crucial.

High-quality HNC treatment requires interprofessional care and a compassionate, person-centered care (PCC) approach, including routine symptom tracking and management (14,15). PCC encourages healthcare providers (HCPs) to listen to patients’ concerns and experiences, empathetically reflect on previously unconsidered values and needs, and co-develop a plan accordingly (16). This requires effective patient–provider communication to empower patients as partners in decision-making (15,17). PCC can improve quality of life and satisfaction with care and mitigate cancer-related distress (15,18).

This project was part of a larger study that trialed a novel education model to promote PCC in HNC care by sharing reports of care experiences in interprofessional staff workshops. The patient-(and caregiver-)as-educator approach was taken (19). Experiences of care were elucidated during focus group conversations with research assistants and HCPs, which were audio-recorded and transcribed. This paper reports on the findings of a Delphi rapid assessment process, undertaken prior to transcript analysis. Focus group facilitators’ perspectives of the most salient elements of focus group conversations were identified and translated rapidly to the HNC team in interprofessional workshops, highlighting actionable ways to improve compassionate PCC. Workshop teachings are described in this report.

Methods

This study was approved by the Research Ethics Board at Sunnybrook Health Sciences Centre (Reference #2057). The Odette Cancer Centre, part of Sunnybrook Health Sciences Centre, is a Regional Cancer Centre ranked amongst the top oncology care sites worldwide. The Odette Cancer Centre houses the third-largest HNC program in Canada, with an interdisciplinary team of specialists across surgery, radiation therapy, chemotherapy, oncology nursing, pathology, speech-language pathology, psychiatry, craniofacial prosthetics, dietetics, and other supportive care services.

Patients and caregivers were recruited from the ambulatory HNC clinic at the Odette Cancer Centre to participate in focus groups. Research assistants approached all patients who met the following criteria, and their caregivers, whenever they were able to do so without disrupting clinic flow: patients (or caregivers of patients) who had undergone treatment for any HNC within the prior 6–18 months, were fluent in English and at least 18 years old. Signs were also posted around the Cancer Centre, prompting interested individuals to contact the research team. Informed consent was obtained by a research assistant. Specific types and locations of tumors were not identified during the recruitment process.

Three focus groups (two patients and one caregiver) of 90 to 120 min duration were conducted in September and October 2019. Focus groups were facilitated by two research assistants with experience in qualitative interviewing (SM and MBK) and three investigators (JE, MJ, and AM) with clinical experience in HNC. Facilitators asked open-ended guiding questions (Online Appendix A), targeting informational or communication needs and experiences of care. Probing questions were used to extract more detail.

The Delphi method is used to derive expert consensus regarding a defined clinical question (20). The consensus was reached using a four-step Delphi process systematically gathering anonymized reports from focus group facilitators on the most salient points of focus group conversations (Figure 1). “Salient,” in this paper, refers to the intensity and frequency (prominence) of the ideas and experiences.

OPEN IN VIEWER

Figure 1.

Stepwise description of Delphi process.

Each facilitator independently listed the most salient elements of the patient and caregiver experience that emerged in the focus groups, from their perspective. Lists were submitted to one facilitator (SM) to maintain anonymity. Facilitators could submit an unlimited number of items. SM consolidated the lists by identifying points most consistently noted and grouping them into categories/themes. At this stage, no points were removed. This revised list was sent back to facilitators for feedback, who were asked whether they agreed with the items included and/or felt items were missing. Feedback was requested regarding categorization and the need for clarification. Facilitators ranked the points in order of salience. Feedback and rankings were sent to SM, who updated the list accordingly. The revised list was sent to facilitators along with any outstanding points not universally listed. Facilitators were asked the same questions again and whether the suggested outstanding points should be included. Outstanding points were only incorporated with majority agreement.

Facilitators universally agreed with the final list of items. Thus, no additional rounds were conducted. After the last round, a question arose regarding the definition of one category, which was resolved through team discussion, rather than anonymous feedback.

Results

Thirteen individuals participated in the focus groups; nine patients (seven male, two female) and four caregivers (four female).

Delphi Process: Each facilitator (n = 5) sent a list of the most salient points discussed in focus groups to SM. Duplicates were consolidated, and points ranked based on frequency, resulting in a list of nine main points, or themes, with several sub-points, or categories. There was an 80% or greater consensus on seven of the nine themes after the first round of feedback. Eighty percent of the group suggested removing the ninth theme and placing it as a sub-point under another theme. Additionally, two facilitators proposed removing another initial point identified as a salient theme, and instead of placing it as a sub-point under another theme. There were also suggestions to add two subheadings beneath two of the themes. In the second round of feedback, all added subheadings were agreed upon (100% consensus) and over 80% agreed to remove the two themes that were not universally agreed upon in the previous round and to list them as categories within other themes. In the final round of feedback, further edits were made to improve readability for reporting purposes with the group consensus.

Content of Focus Group Discussions: Seven main points were identified, with several sub-points (described below, summarized in Figure 2).

OPEN IN VIEWER

Figure 2.

Seven salient themes (with sub-categories) were discussed in focus group conversations regarding patient and caregiver experience of head and neck cancer.

Receiving Relevant Information at the Most Appropriate Time: Many participants reported receiving a large amount of information at their first oncology appointment. This was described as overwhelming and difficult to navigate. Participants voiced a preference to receive information when relevant, rather than all up-front.

Diversifying Information Provision: Patients reported receiving most information in the form of pamphlets, describing this as unengaging and difficult to organize. Participants recommended providing information in different forms of media.

Setting Realistic Expectations: Participants reported appreciating honest communication about their condition, as this allowed them to have “positive but realistic” expectations. Most participants emphasized not wanting things to be “sugar-coated.” 2.

There were two most salient categories in the theme of burden related to symptoms and treatment side effects :

Differing Satisfaction With Symptom Management and Supportive Care: Some participants reported that their symptoms were managed well and felt satisfied with their care. Others said they would have benefitted from supportive care services or that the most appropriate resource was not offered. Some participants were offered, but did not access or were not satisfied with, supportive care services, feeling that others might need the services more or that attending would be an additional burden. 3.

Importance of Social Support

At times, participants felt that their quality of care was compromised. At the systems level, examples were related to respect for time, for example, long wait times and last-minute appointment changes), having needs addressed (e.g., requests on the nursing line not answered), and coordination of care. At the provider level, incidents of incorrect tests, referrals, and diagnoses were distressing to participants. 5.

There were two main categories in Caring for the person, not just treating cancer :

Human Connection and Trust: Participants valued providers taking the time to actively listen to their concerns and respond to questions. Some participants reported appreciating when clinicians would smile, keep the mood light, make eye contact, and use humor and physical touch. Participants highlighted HCPs who were perceived as confident and capable, as well as caring. 6.

Social and Emotional Impact of HNC

By chance, three out of four participants in one focus group were diagnosed with HPV-related HNC, so this was a dominant topic of conversation. Patients and caregivers with HPV-related HNC expressed unique informational and psychosocial needs. Some patients chose not to share their diagnosis with friends and family, reporting feelings of guilt and stigma. Similarly, caregivers reported guilt related to their partners’ HPV, feeling that the cancer was “their fault.” Most reported receiving minimal information, expressing a need for more information about the HPV diagnosis, particularly concerning intimacy and broader implications for their health and their partners’ health.

Discussion

Focus groups illuminated many factors impacting the patient and caregiver experience of care, resulting in actionable recommendations for improving future PCC. PCC promotes flexibility in the timing, amount, and method of information-sharing based on patient preferences, addressing individual informational needs thereby increasing their understanding and information retention.

Patient-Centered Communication improves informed medical decision-making, patient satisfaction, treatment adherence, and medical outcomes (21). People with cancer, specifically HNC, often want maximal information about their illness and prognosis (22). Unfortunately, many have inaccurate illness understanding (23). HCPs often underestimate informational needs and struggle with communicating poor prognosis or “bad news,” (24) possibly due to insufficient training (22). The present study revealed a desire for HCPs to be “positive but realistic” when discussing their prognosis and long-term burden of treatment. Patients appreciated a sense of optimism but wanted accurate expectations for preparing ahead.

Many patients struggle to navigate information about their illness, sometimes due to emotional barriers such as feeling anxious or overwhelmed at the prospect of integrating information, or scalable barriers related to the constant accumulation of information (25). When receiving excessive information up-front, patients face temporal barriers including insufficient time to learn and organize the information (25). To address these barriers, participants in the present study recommended ongoing information provision, highlighting what is currently relevant to the individual. This might maximize patients’ ability to benefit from the information provided, thus overcoming functional barriers (25).

Information can also be personalized according to the individual's type of cancer. Participants with HPV-related HNC voiced the desire for more information about HPV. Oncologists may lack sufficient training in providing HPV-related counseling (26). Confusion and misinformation can contribute to stigma, negative emotions, poor illness understanding, implications for relationships and intimacy, and decreased compliance with follow-up and disclosure of HPV status to partners (27). Distress about being HPV-positive decreases when informational needs are met (28). Finnigan and Sikora (29) developed a guideline for counseling patients with HPV-related HNC, addressing major areas of patient concern. Referrals to other support services can help address this issue (27).

Participants also highlighted informational needs post-treatment, when appointments are less frequent. Long-term contingency planning and expectation-setting may benefit patients, given the lasting impact of some HNC-related side effects (i.e., impaired eating, speaking, and tasting). Though HCPs may be wary of discussing anticipated side effects while patients are doing “well,” reminders can help patients prepare for harsher symptoms toward the end of, or after, treatment. Patients may appreciate being reminded of how to access support and information related to side effects.

Finally, participants recommended diversifying the form of information provision, that is, using different media (including discussion), rather than mostly written pamphlets. Discussion may help to increase the quality of patient-centered communication, by allowing patients to ask questions and request further information, where needed. This would overcome barriers that prevent patients from addressing their own informational needs, reported in this study and the literature, such as feeling that there is not enough time in the clinic to ask questions, reluctance to bother HCPs, or believing that HCPs will not be able to provide answers or solutions (30).

Understanding the individual’s psychosocial needs and preferences for support might improve the quality of PCC and system efficiency, as individuals reported differing needs and desires related to psychosocial support at different points in their illness trajectory. Ongoing assessment of patients’ psychosocial needs over the course of treatment with judicious timing of referrals to psychosocial resources might increase patients’ likelihood of accepting such support. Many patients are reluctant to share emotional symptoms and psychosocial issues are under-discussed in cancer care (31), as confirmed in this study.

Barriers to discussing distress in appointments and accessing specialized support include lack of psychosocial evaluation, time, resources, and awareness of helpful resources (32–34). Methods of increasing psychosocial care and greater resource utilization might include providing simple pathways of referral to oncologists, reminding teams to use symptom screening (35,36), increasing knowledge of resources, and appreciation for the importance of psychosocial support in cancer treatment and recovery.

Not all patients reported a desire for formal support; common reasons included already receiving informal help elsewhere or a preference for self-management of distress (37). Physicians should focus interventions on those willing to accept support (37). Both patients and loved ones reported positive views regarding peer support. Some described benefitting from connecting with others in waiting rooms or sharing their experiences with friends with previous cancer experience and receiving mentorship from their peers. Others noted the desire for peer support through their treatment, stating a formal peer support program would be a useful addition to the cancer centre. Despite the large interest in peer support expressed by participants, previous research shows low engagement in peer support when available (38,39). Barriers to engagement in peer support include stigma, lack of awareness, limited time, and the emotional burden of discussing disease (40), though, peer support might actually help to mitigate stigma. Most peer support programs tend to reach middle-aged, educated women without advanced cancer (41).

Peer supporters can help to inform and educate patients and their loved ones, reduce anxiety, increase a sense of empowerment, and may improve quality of life and coping (42–45). Some studies showed no change in health-related quality of life or distress (46,47), possibly reflecting different patient preferences regarding peer support and its delivery (e.g. group or individual). This underlines the importance of HCPs having some time to listen to and understand individuals' needs, preferences, and coping methods, so that they can help patients and their loved ones access support most suited to them. HCPs need to be aware of available supports, including peer support to recommend the most appropriate resources.

Feeling cared for as a person, not just a patient, was described by many when discussing positive experiences of care. On a hospital level, this was characterized by streamlined and coordinated care, considering the individual's schedule, making treatment accessible and convenient. On an individual provider level, participants appreciated those who acknowledged their unique challenges and provided tailored “whole-person” care. Participants defined compassionate care as genuine and authentic—characterized by the perception that HCP truly cared for them as a person. Some actionable ways for HCP to show compassionate care identified during this study include clinicians taking time to introduce themselves and connect to patients and caregivers by sitting down, making eye contact, listening attentively, asking follow-up questions, and using physical touch (e.g., hand on the shoulder). Understanding compassionate care from the patient and caregiver perspectives can enable HCPs to provide high-quality care. Not only does compassionate care decrease distress and improve patient satisfaction, but also has benefits for clinicians such as increased job satisfaction and retention (48,49).

Limitations and Future Research

There are several limitations of this study. Firstly, the experiences described may not be generalizable due to the limited sample size. However, salient elements of the HNC experience identified echo those noted in related studies. The sample size and methodology taken also did not allow a thorough between-group comparison of patient and caregiver reports. Future work with larger sample sizes might consider exploring the similarities and differences in experience and needs between these groups. Similarly, the sample included a large proportion of patients (and caregivers of patients) with HPV-related HNC, potentially limiting the generalizability of some findings to other sub-populations. Future research exploring the specific needs of varying sub-populations could lead to the creation of guidelines for personalizing care based on the specific form of an individual’s cancer.

Finally, there is a risk of bias in the Delphi method, due to facilitators’ previous knowledge and experiences. Methodological guidelines for the Delphi Method were followed (50); facilitators submitted feedback anonymously. One facilitator consolidated and refined all feedback to limit group bias. Finally, the panel of facilitators came from different professional backgrounds and had different levels of experience or familiarity with HNC. Future work will report on the similarities and differences between findings of the Delphi process versus the thematic analysis of focus group transcripts.

Conclusions

Our findings present seven salient themes capturing the experiences of HNC care for patients and their caregivers. Listening to the perspectives of patients and caregivers allowed us to identify three overarching recommendations to improve patient care—using patient-centered communication, seeking to understand the individuals’ psychosocial needs and preferences for support, and aiming to make patients feel cared for as a person, not just a patient—all of which relate to the practice of PCC. The experiences of patient and caregiver participants in our study presented in the context of a review of relevant literature establishes a solid foundation for proposing concrete strategies to address these areas of concern during cancer care.

Supplemental Material