One-on-one, digitally recorded qualitative interviews with family caregivers, patients with stage I to IV cancer, and oncology clinicians and administrators were conducted in this formative evaluation study to gather feedback on the content, structure, and delivery of the FamilyAWARE concept. Participants were recruited from a large academic cancer center in the US South between January and July 2025. Interviews were conducted by telephone, in person, or by HIPAA (Health Insurance Portability and Accountability Act)–compliant Zoom. This report followed the COREQ (Consolidated Criteria for Reporting Qualitative Research) guidelines and its attendant checklist [27].

This study was approved by the institutional review board (IRB) at the University of Alabama at Birmingham (IRB-300007406). All participants provided informed consent and were paid US $50 as an incentive for participating in the study. The study team established and enforced measures to protect against unauthorized access, use, and disclosure of personal health information to ensure confidentiality and data integrity. Participants’ names were not transcribed, and all other identifiers were removed.

Family caregivers were eligible if they were aged ≥18 years, self-endorsed or identified by their care recipient as “an unpaid relative or friend who knows them well and who provides regular support due to their cancer,” caring for a patient with stage I to IV cancer receiving active treatment, and English speaking. Caregivers were excluded if they self-reported severe, untreated mental illness (eg, schizophrenia, bipolar disorder, or major depressive disorder), dementia, active suicidal ideation, uncorrected hearing loss, or active substance abuse. Patients were eligible if they met the following inclusion criteria: aged ≥ 18 years and diagnosed with a stage I to IV cancer (including hematologic cancers) and receiving active treatment (ie, chemotherapy, targeted therapy, or immunotherapy). Patients were excluded if their electronic health record (EHR) documented active severe mental illness (eg, schizophrenia, bipolar disorder, or major depressive disorder), dementia, active suicidal ideation, uncorrected hearing loss, or active substance abuse. Oncology clinicians—including (but not limited to) physicians, physician assistants, nurse practitioners, nurses, social workers, psychologists, behavioral health counselors, navigators, and administrators—had to have at least 1 year of experience working with patients with cancer and their caregivers. A targeted sample size of 12 to 25 individuals (per constituency) was targeted to achieve saturation [28,29].

Purposive sampling was used to ensure representation across cancer diagnoses (patients and caregivers) as well as professional backgrounds (clinicians). A clinical research coordinator (EO) invited patients receiving active cancer therapy and their caregivers to participate, while clinicians were recruited via email based on their engagement with the cancer center’s preexisting RSM initiative [21,30-33].

Interviews were conducted by trained study staff using an interview guide adapted from the team’s prior qualitative formative evaluation work (available upon request) [34,35]. Before the interview, participants received a 1-page outline of the FamilyAWARE intervention concept (Figure 1), either in hard copy or by email. During the interview, participants were oriented to the outline. The FamilyAWARE intervention concept and outline were grounded conceptually on the theory of dyadic illness management [17] and social support effectiveness theory [18-20] and structurally developed based on social support and symptom management coaching content from the Project Cornerstone and CASCADE family caregiver interventions developed by the first author (JNO) in prior work [23,25,36-39].

The tentative intervention outline was proposed to consist of two major components: (1) coaching sessions about how to best help the patient manage symptoms and (2) access to a secure online web page that summarized their loved symptoms (as tabulated from the patient’s ePRO symptom reports) and provided tips on how to help. The coaching sessions would comprise 3 semistructured, weekly coaching sessions lasting 20 to 30 minutes each. Sessions would focus on how to be an effective partner with your loved one in managing symptoms, strategies for tracking symptoms and medications, and orientation to a web-based dashboard that would summarize your loved one’s symptoms. The secure online web page, or “dashboard,” would be designed to be user-friendly, displaying the patient’s most recent symptom reports and providing personalized symptom management tips based on the patient’s symptoms. Caregivers would receive a notification when new symptom reports became available. Participants were informed that any aspect of the intervention could be modified and were asked about their overall impressions; the topical content, number, and length of coaching sessions; the accessibility and usability of the dashboard; and strategies to attract and engage patients and families with this type of intervention. All interviews were transcribed by a professional transcription service (Landmark Associates, Inc). Deidentified transcriptions were uploaded into NVivo (version 14; Lumivero) qualitative data analysis software.

Following a thematic analysis approach [40,41] aligning with our prior work [34,35], an interactive series of steps was taken to analyze the data, including (1) immersing in the transcribed interviews through careful reading, (2) open coding, (3) generating initial overarching themes, (4) using within- and across-case comparison [42] to further refine and define themes, and (5) having additional members of the study team (RB and MLS) review audit trails of the developing themes to assess fit of the raw data. For steps 1 and 2, the principal investigator (JNO), who has extensive experience and expertise in family caregiving and qualitative research, carefully read all transcripts and conducted line-by-line open coding. Following initial coding to reduce the data, codes and their associated data were grouped into broader thematic categories in step 3, using matrices to support both within- and across-case comparisons. To refine the preliminary themes as part of step 5, additional members of the study team (RB and MLS) reviewed the audit trail of supporting quotations and evaluated the coherence and comprehensiveness of the themes in relation to the entire corpus of supporting data.

Qualitative rigor was guided by the criteria of credibility, transferability, dependability, and confirmability proposed by Lincoln and Guba [43]. Credibility was supported through purposive sampling across family caregivers, patients, and oncology clinicians; continued recruitment until thematic saturation was reached; iterative within- and across-case comparisons; and review of the developing themes by multiple members of the study team. Transferability was supported by detailed reporting of the study setting, participant characteristics, proposed intervention components, and example participant quotations. Dependability was addressed through a systematic, documented analytic process and maintenance of an audit trail describing coding decisions and theme development. Confirmability was supported by grounding themes in the underlying participant data and having additional investigators review the audit trail of quotations and assess the coherence and comprehensiveness of the themes in relation to the full data corpus.

Participant characteristics are shown in Table 1. Family caregivers (n=20) had a mean age of 59.2 (SD 12.6) years and were mostly female (n=14, 70%), White (n=12, 60%), employed full or part time (n=12, 60%), married or living with a partner (n=13, 65%), and the husband or wife of the patient (n=8, 40%). Most caregivers provided support to patients 4 or more days a week (17/20, 85%) for 3 or more hours a day (13/20, 65%). Patients (n=20) with a wide range of cancers had a mean age of 57.9 (SD 12.3) years and were mostly female (n=15, 75%), White (n=12, 60%), and employed full or part time (n=10, 50%). Clinician participants (n=25) were mostly female (n=21, 84%) and included physicians (n=6, 24%), nurses (n=8, 32%), or nonclinical navigators (n=5, 20%).

Main themes are presented in the following sections. Table 2 depicts these main themes with additional example quotes.

Results from interviews with 20 family caregivers, 20 patients with stage I to IV cancer, and 25 oncology clinicians showed overall enthusiasm for the caregiver-focused symptom partnering intervention concept and yielded valuable insights into its design. On the basis of some of the team’s prior family caregiving intervention work [23,38], the initial FamilyAWARE intervention concept presented in interviews would integrate one-on-one coaching to train caregivers on how to partner effectively with patients, track symptoms and medications, and access a web-based dashboard summarizing their patient’s symptoms with tailored recommendations. Findings endorsed several features of the presented concept, including the coaching session content, the length and number of sessions, caregivers’ access to patient symptom data (provided that patient permission was obtained), and the tailored recommendations for providing symptom support. However, participants were more critical of other features and provided insight into additional considerations for refinement.

FamilyAWARE is intended to complement rather than replace validated ePRO measures and existing RSM systems. These systems have demonstrated clinical benefits and are increasingly integrated into EHR and patient portal workflows [13-16,30-33,44]. Although systems may permit authorized caregiver access to patient information [45], access to symptom data alone does not necessarily provide caregivers with structured coaching, tailored interpretation, or actionable guidance about how to support the patient. The proposed contribution of FamilyAWARE is therefore a caregiver-focused layer that translates existing patient-reported symptom information into permission-based coaching and practical recommendations. Future development should prioritize integration with established ePRO and EHR infrastructure rather than creation of a duplicative symptom-reporting platform.

One feature of the FamilyAWARE concept that participants were the most apprehensive about centered on individuals’ digital literacy, access, and willingness to learn and navigate a web-based dashboard. This concern aligns with a growing body of literature, including reports from the National Academy of Medicine [46-50], highlighting the “digital divide,” which can undermine engagement among those who may benefit most, including older adults, rural populations, and racial and ethnic minority groups. This suggests that an alternative to a web-based platform is needed for sharing patient symptom reports and recommendations with caregivers. One possibility mentioned by participants was having symptom reports accessible via smartphones, potentially using text messaging. Mobile phone ownership is far more widespread, with smartphones now owned by 90% of the US population [51], making these devices more available and readily accessible for daily use than desktop or laptop computers. Furthermore, a text-based mobile platform typically requires less digital literacy compared to complex web navigation, potentially lowering barriers to caregiver access and engagement. It may also be helpful to view FamilyAWARE not as a single required solution but as 1 option within a suite of caregiver support tools, recognizing that different caregivers will need and prefer different formats.

While first impressions of FamilyAWARE were generally positive across participants, with caregivers expressing the strongest enthusiasm, a critical factor highlighted by participants to optimize the initial “buy in” and engagement by caregivers was framing the intervention in a way that emphasizes its value to their caregiving role and the support they provide to patients. Participants repeatedly described caregivers as operating under considerable time, task, and psychological demands, making caregivers more selective and hesitant about the adoption of new tools and resources. Thus, uptake was believed to be more likely if FamilyAWARE was presented as a program to help optimize their partnership with patients and to streamline existing symptom support responsibilities and not simply add another task to a caregiver’s daily routine. This aligns with prior research showing caregiver engagement in research to be influenced by perceived relevance and utility [52,53]. Designing FamilyAWARE to be a flexible resource that can be used more or less to fit a caregiver’s needs may further reinforce this framing and avoid the perception of a one-size-fits-all approach. Implementation of FamilyAWARE will thus need to consider this framing as a mechanism to promote intervention fit, using messaging and onboarding processes that connect the intervention’s features with caregiver priorities and reduce, rather than add to, their workload.

A final feature of the intervention concept emphasized by participants, particularly clinicians, was the essential need for symptom-specific guidance on what to expect and when to seek additional assistance and guidance. Caregivers talked about uncertainty around what to expect as “typical” for patients’ symptoms and desired to have more upfront understanding about what to prepare for. There was also a desire for alerts built into the platform that might flag concerning trends or symptoms that might require special intervention or an escalation in support. This approach is consistent with prior research in RSM, which suggests that symptom reports and automated alerts can improve patient symptom outcomes, quality of life, and health care use [13-15,44].

While our study reports critical feedback from caregivers, patients, and clinicians on the FamilyAWARE intervention concept, it is not without limitations. First, because this was a qualitative study, the sample size was small and not representative of the larger population of patients and their families. Moreover, participants were recruited from a single site, further negating generalizability. Second, it is possible that a larger sample size may have increased the amount and depth of feedback acquired. While saturation was obtained, we cannot eliminate this possibility. However, the next phase of this research will involve manualizing and pilot testing the FamilyAWARE intervention with a larger sample using an actual prototype, allowing for additional feedback on its content, format, and delivery and further refinement of the intervention. Third, because study participation and interviews were limited to English-speaking individuals, the findings may not reflect the perspectives of people who require communication in other languages. We also did not formally assess health literacy or language and health literacy concordance, limiting our ability to examine whether intervention preferences varied according to these characteristics.

This qualitative formative evaluation study found strong interest among caregivers, patients with stage I to IV cancer, and oncology clinicians for a proposed intervention concept called FamilyAWARE that combines one-on-one coaching with online dashboard access to patient-reported symptom data. Participants valued the proposed coaching sessions and easy-to-understand symptom guidance, provided that the program minimized burden, was mobile-friendly or mobile-based, and accommodated varying levels of digital literacy. Engagement was viewed as most likely when the program was framed as adding value to, rather than adding tasks to, the caregiver role. Findings will directly inform refinements to FamilyAWARE’s content, delivery, and framing for pilot testing, with the goal of enhancing caregiver effectiveness and supporting patient outcomes in cancer symptom management.