This qualitative study used semistructured interviews to explore how YA cancer survivors find, access, and engage with online cancer communities. The study aimed to understand both the pathways through which participants discovered these communities and the barriers they encountered in participating or sustaining engagement over time.

Purposive and snowball sampling was used to recruit participants. To be eligible for the study, participants needed to (1) receive a diagnosis of cancer at the age between 18 and 39 years, (2) be between 18 and 39 years at the time of the interview, (3) have experiences using social media platforms, and (4) be able to participate in a videoconference interview. The age range was based on the National Cancer Institute’s definition of adolescent and young adult (AYA) cancer survivors [5] and was selected to capture the relational and developmental contexts of young adulthood, a life stage marked by key social transitions and identity formation that shape survivorship experiences [6].

The research team developed a flyer that they asked nonprofit organizations dedicated to YA cancer to post on their social media accounts. Cactus Society and Livestrong Cancer Institute, 2 US-based organizations supporting AYA cancer survivors, shared the flyer on their social media accounts. The study flyer contained study aims, the principal investigator’s contact information, and a QR code of intake survey on Qualtrics. The principal investigator (QC) followed up with each potential participant to schedule Zoom interviews and gather verbal informed consent.

All interviews followed the same interview guide with some variations based upon the flow of conversation between the participant and interviewer (see Multimedia Appendix 1 for the interview guide, including prompts and questions). The interview guide was developed by the research team. It contained questions related to the ways YA cancer survivors connected to peer survivors on social media, including positive and negative experiences during the process. The interviews were conducted in a one-on-one setting, free from the presence of any nonparticipants. Prior to each interview, participants were instructed to join from a private location and to ensure that no nonparticipants were present or able to overhear the conversation. Interviews were video recorded and transcribed by Zoom. Transcripts were made available for participant review upon request. The interviewer (QC) read through transcripts to ensure the quality of transcription and kept field notes that captured the main points of conversation after each interview.

Reflexive thematic analysis is a method for identifying, analyzing, and reporting patterns within data, defined by Braun and Clarke [24]. All the transcripts were deidentified before importing into MAXQDA (VERBI Software, 2021), a qualitative software program for data coding and comparison.

Two members of the research team, QC and LZ, independently coded all transcripts using an inductive, consensus-driven approach. The data coding approach was inductive. Codes and theme development were directed by the content of the data, rather than being theoretically driven [24]. Both coders initially reviewed a subset of transcripts to develop a preliminary codebook, which was refined through iterative discussion. Each transcript was then coded separately by both coders, followed by comparison and discussion to reconcile differences and reach agreement on code application and interpretation. Discrepancies were resolved through dialogue until full consensus was achieved, ensuring analytic consistency and rigor.

Following the six-phase process of thematic analysis, QC and LZ began with a process of (1) data familiarization, in which QC noted initial analytic observations about each data item and the entire dataset by reading the transcription and field notes. Two transcriptions were coded to (2) generate initial codes and develop the first-version codebook with columns of code category, code name, definition, eligibility, and example based on the initial codes. QC and LZ coded the other 2 interviews and revised the codebook accordingly by adding new codes, refining code’s eligibility criteria, and adding second-level codes to group the existing codes. After the codebook was finalized with all 12 interviews, the data were organized into a matrix with cases and second-level codes. QC and LZ (3) searched for themes by reading and identifying the similarity in data across cases in the matrix. The 2 authors (4) reviewed the themes and gave (5) definitions and names to the themes. All authors reached consensus regarding the themes’ consistency. The last phase was to move from semantic description to interpret the significance of patterns and their broader meaning and implications for the (6) final report.

Steps to ensure rigor were based on the trustworthiness criteria of data analysis posited by Lincoln and Guba [25]. The criteria involved establishing credibility, dependability, confirmability, and transferability. Throughout the research process, the first author reported the coding process and reflection of the transcription with the research team to assure the credibility of the analysis. The research team provided consultation and guidance on the naming of codes and themes to ensure the accuracy of language. The first author kept an audit trail (Koch [26]) to record each decision made with modifying the research question, finalizing the data analytic approach, and coding and analyzing the pattern to improve the dependability of the finding. Qualitative software was used, as it permits greater audibility and hence increases the reliability of the data analysis process.

Confirmability was sought by regular review and discussion of the codes with the research team by discussing reflexivity and how the interactive nature of the interview process influenced data collection. Member checking was not conducted, as this study employed a reflexive thematic analysis approach in which meaning is understood as coconstructed through researcher interpretation rather than verified by participants. Following Braun and Clarke’s distinction, traditional member checking assumes the existence of a single, verifiable truth and seeks to “correct” researcher subjectivity—an assumption that sits conceptually at odds with reflexive thematic analysis, where researcher subjectivity is viewed as an analytic resource, not a bias to eliminate [27]. Consistent with the interpretivist assumptions of reflexive thematic analysis, data saturation was not used as a criterion for determining sample adequacy because Braun and Clarke [28] advocated for achieving meaning through researcher interpretation.

Institutional review board approval was granted for this study by the University of Texas at Austin (protocol STUDY00003004). All study procedures complied with institutional and national research ethics standards for research involving human participants. Participants were informed of the study purpose, confidentiality protections, the benefits and risks of participation, and their rights to withdraw at any time. Prior to participation, all participants provided verbal informed consent to take part in the study and to be video recorded. Participation was voluntary, and participants were reminded that they could decline to answer any question or withdraw from the study at any time without penalty. All data were deidentified during transcription and analysis. Identifying information such as names or specific personal details was removed or replaced with pseudonyms. Digital recordings and transcripts were stored in a secure and access-restricted folder on UT Box, accessible only to the research team. Each participant received a US $20 Amazon electronic gift card, sent via email upon completion of the interview, as compensation for their time and contribution to the study.

The research team represents a diverse group of scholars and practitioners whose social identities, disciplinary backgrounds, and professional experiences shaped the design and interpretation of this study. QC (Asian woman, Assistant Professor of Social Work, and a PhD candidate at the time of the study), ED (White woman, Professor of Communication), LN (Black woman, Assistant Professor of Social Work), and BJ (White woman, Professor of Social Work) are seasoned qualitative researchers with a PhD degree and extensive experience studying psychosocial oncology, communication, and health equity among AYA cancer survivors. LZ (Asian woman, social work clinician, and doctoral student), the second coder of the transcript, brought both clinical practice experience and methodological expertise in qualitative research. The interviewer QC’s interest in this topic stems from her professional experience in psychosocial oncology and her commitment to addressing disparities in access to cancer support as well as the rising use and impact of communicative technology during the COVID-19 pandemic. She acknowledges that her background as a social work researcher and her own positive experiences with online communities as an immigrant may have shaped assumptions about online support and digital disparities. As someone who is not a cancer survivor herself, she remains mindful of the potential distance between her professional perspective and participants’ lived experiences. QC had limited prior relationships with participants. One participant was a personal friend of QC, while all others had no prior relationship with the interviewer. Participants were informed about QC and the study through the introductory phone calls or emails and the consent form, which described QC’s role as the principal investigator and explained the study’s purpose, goals, and procedures. No additional personal or professional information was shared beyond what was stated in the consent materials.

A total of 20 potential YA participants who filled out the screening questionnaire were all eligible for the study. Of these 20 YAs, 8 did not respond to the principal investigator’s emails or phone calls, and 12 consented and were recruited to participate in the interviews, yielding an enrollment rate of 60%. Figure 1 presents the recruitment and analysis process flowchart. Semistructured individual Zoom interview data were collected from September 2022 to February 2023, ranging from 39 to 110 minutes (mean 67.2 min, SD 21.3 min). No repeated interview was carried out, and no participants were requested to review or edit their transcripts. The demographic information of participants is presented in Table 1. Participants had a mean age of 32 (SD 5.08; range 24‐39) years and an average of 3 (SD 2.98; range 0‐11) years since diagnosis. Most participants were identified as female (n=9, 75%). The majority were non-Hispanic White (n=7, 58%), followed by Hispanic (n=2, 17%) and Asian (n=2, 17%), and 1 (8%) identified as multiracial. Over half (n=7, 58%) of participants held a graduate degree, while 4 (33%) had completed some college or a college degree, and 1 (8%) had completed high school. In terms of employment, 6 (50%) participants were employed full-time, 2 (17%) part-time, and 4 (33%) not employed. Participants resided in 8 different states. Most participants resided in suburban areas (n=8, 67%). Cancer diagnoses among participants varied, with leukemia being the most common (n=3, 25%), followed by Hodgkin’s lymphoma (n=2, 17%), testicular cancer (n=2, 17%), and individual cases of non-Hodgkin’s lymphoma, colon, brain, soft tissue, and uterine cancers (each n=1, 8%). At the time of the interview, 9 (75%) participants had completed treatment, while 3 (25%) were still in active treatment.

This study explored 2 central questions: (1) how YA cancer survivors locate and access online cancer communities and (2) what challenges they encounter in sustaining engagement. The findings revealed a range of entry points that survivors used to discover supportive online spaces, as well as structural, interpersonal, and emotional factors that limited their continued participation. While participants represented a range of ages within the YA category (24‐39 y), their approaches to engaging with online cancer communities and the obstacles they experienced were broadly similar. No age-related patterns were observed in terms of search strategies, preferred platforms, or engagement levels.

The results are presented in 2 sections, corresponding to each research question. The first section, Pathways, included 5 themes describing how YA cancer survivors found online communities: (1) search, connect, and join/build, (2) community hubs, (3) third-party referrals, (4) algorithm recommendations, and (5) online interest-based groups. The second section, Roadblocks, outlined 5 themes that capture barriers to sustained engagement: (1) platform fragmentation and digital literacy, (2) group boundaries and hierarchies, (3) lack of representation, (4) emotional overload and engagement fatigue, and (5) cyberbullying. Figures 2 and 3 depicted 2 coding trees of Pathways and Roadblocks to illustrate the codes identified and themes emerged in the analysis. Table 2 presents each theme along with its definition and representative participant quotes.

YA cancer survivors accessed online communities through a range of entry points that reflected both deliberate strategies and serendipitous discovery.

While online communities offered vital spaces of support, YA cancer survivors reported a number of challenges that shaped their participation.

This study provided timely insights into how YA cancer survivors navigated the increasingly complex landscape of online cancer communities. It identified 5 key pathways to discover online communities and 5 major barriers in finding and sustaining engagement. The findings revealed that survivors’ participation was influenced not only by personal agency and constraints but also by systemic factors such as platform accessibility and usability. The findings also further illuminated understudied interpersonal dynamics, such as emotional labor and shifting hierarchies and boundaries within communities, that affected survivors’ sustained involvement in online communities. By centering YA cancer survivors’ lived experiences in exploring online communities, this research advanced a more nuanced understanding of how to foster inclusive, meaningful, and sustainable online support communities.

This study uncovers new forms of digital space navigation and structural inequity that shape survivors’ ability to benefit from online support. Our findings showed that with more online platforms available, YA survivors joined or formed communities across different platforms and often switched between them. While new features such as algorithmic recommendations, hashtags, stories, and group chat channels allowed them to find or form online cancer communities more easily, the benefits of these tools were not experienced equitably. Participants’ capacity to navigate and engage effectively was strongly influenced by their familiarity with functions of different platforms, many of which were not accessible to all users. For instance, immigrant survivors originally from countries where platforms like Facebook and Instagram are banned often lacked exposure to dominant US platforms. It demanded additional effort to learn and transition to a different platform solely for the purpose of accessing online support. Furthermore, limited access to certain social media platforms among users from specific cultural backgrounds results in fewer online support communities offering inclusive and culturally relevant content in their native languages. This disproportionately affected YA of color. These gaps in representation were compounded by algorithmic patterns that reinforced misinformation and unsolicited advice, disproportionately affecting those with lower health literacy or limited English proficiency. Together, these findings reveal how structural digital inequities shaped YA survivors’ capacity to benefit from online support.

This study reveals that YA survivors’ engagement with online cancer communities was emotionally dynamic and relationally fluid. Their interactions were marked with uncertainty and change. Navigating online cancer communities involved ongoing assessment of the relevance of the communities, negotiation of personal or group boundaries, and weighing between benefits and the risks. Some participants moved fluidly between communities or platforms as new communities emerged and migrated, while others withdrew temporarily or permanently when the emotional toll became too high. Rather than serving as a stable or consistently supportive resource, online cancer communities could also generate emotional distress, uncertainty, and exposure to unsafe or overwhelming virtual environments.

This study extends prior digital health disparity research on YA survivors’ online support by shifting attention from individual constraints in digital literacy to structural inequities in virtual spaces. Prior research has largely emphasized individual-level factors such as education, socioeconomic status, or their social capital as the main drivers of digital divide in access and utilization of online resources [29,30]. Our study reveals that digital disparities persisted even among digitally savvy YA survivors, driven by the nature of online platforms. Participants’ platform preferences were shaped by the habits as well as the accessibility of apps and websites of their families and social networks [30]. One prior study has similarly identified representation barriers, which noted that YA survivors often avoided older platforms such as Facebook, where most established cancer support groups exist, due to perceived misalignment with their peer networks and identity [13]. Familiarity with specific platforms, access to culturally relevant content, and the design of digital infrastructures emerged as key determinants equally impacted YA. This shift from individual to systemic explanations underscores the importance of improving platform accessibility and usability.

This study also advances existing literature by revealing how interpersonal dynamics within online communities could introduce more emotional distress and uncertainties besides support. Consistent with the prior findings, our study found that YA cancer survivors had strong needs of connecting to peer survivors [31] and they would search online proactively to find peer survivors [32]. Our study pointed out that, however, engagement was not always self-initiated; algorithms or others’ suggestions often prompted participation, leading to interpersonal strain. Prior work has well documented the benefits for YA cancer survivors involving online cancer communities, emphasizing information exchange, social support, and empowerment [31,33,34]. Risks such as exposure to distressing content or misinformation have also been acknowledged, though often discussed as isolated, individual experiences [35]. One prior study has examined the dynamics of social media engagement among YA cancer survivors, but focusing primarily on individual motivation and preferences [32]. Our findings expand these understandings by showing that group-level factors such as changing boundaries, evolving hierarchies, and the associated emotional labor directly influence engagement and withdrawal. By framing these experiences as relational and evolving, this study highlights that community organizing and group moderation is necessary for virtual spaces.

This study has several limitations. First, our sample included only participants who had internet access and prior experience using social media. It remained unclear whether YA cancer survivors without such access or experience would be open to, or able to engage with, online cancer communities. This inclusion criterion was necessary to ensure that participants could reflect meaningfully on their experiences navigating online communities. However, it may have limited the perspectives of YA cancer survivors with limited digital access or literacy or who lack interests in online cancer communities. Future studies should consider various recruitment strategies (eg, phone interviews or community-based outreach) to capture the experiences of survivors who have not been involved with any online cancer communities.

Second, the sample lacked representation from African American participants. Prior research indicated that African American cancer survivors were underrepresented in online support groups and less likely to use the internet for health information compared to their White counterparts, due in part to digital access barriers, differing help-seeking preferences, and lower trust in online health sources [36,37]. Although we recruited through multiple nonprofit networks serving diverse populations and encouraged participant referrals, our sample may have unintentionally excluded survivors not affiliated with such organizations or those relying on offline or culturally specific support systems. Future research should employ more inclusive recruitment strategies, including outreach through hospital registries, partnerships with African American survivor networks or community centers, and targeted social media campaigns or accounts that have higher number of African American followers. Addressing these structural and digital inequities is essential to advancing equity and representation in digital cancer survivorship research.

Future intervention research and clinical practice should focus on improving the accessibility and sustainability of online cancer communities through both structural support and individual capacity-building.

At the system level, oncology social workers have long played a critical role in connecting patients with fragmented resources [38], and our findings further confirmed this essential function. However, many social workers, particularly those outside of large cancer centers, were not well-versed in the availability of online cancer communities and virtual support groups for minority YA survivors on various scattered platforms [31]. While social support from family, peers, and health care providers can facilitate access to online communities, the active involvement of health care professionals or trained moderators in online health communities enhances trust, reduces the spread of misinformation, and mitigates risks such as cyberbullying and communication breakdowns [39,40]. Expert-led moderation has been associated with improved emotional safety, greater engagement, and more supportive peer interactions [41]. This underscores the importance of digital literacy training with health care professionals to strengthen their competencies in navigating and moderating different online cancer communities, so they can more effectively refer, prepare, and support YA survivors.

At the individual level, private or peer-led groups typically rely on YA survivors themselves to establish norms and maintain boundaries, placing added emotional and relational burdens on those already coping with cancer. Our findings indicated that many YA survivors not only passively join a group but actively take part in building or contributing to their own communities. One promising approach is to train YA cancer survivors who are interested in leadership roles in online communities on skills to identify harmful content, establish group norms, and manage conflict. Additionally, while the prevalence of cyberbullying and emotional risk in online communities has been documented [42], few interventions exist to prepare survivors for these challenges. Education around safe storytelling, self-care, and managing online engagement fatigue should be integrated into survivorship support programming [35]. This includes guidance on when to disengage, how to set personal boundaries, and how to respond to toxic interactions.

This study highlights that YA cancer survivors navigate online communities through multiple pathways but face significant barriers to sustained engagement, highlighting the need for provider training and peer moderator support. To foster safer and more inclusive online environments, health care providers should be trained to guide survivors toward reliable support resources; community moderators should be trained to manage emotional labor and maintain healthy group boundaries; and YA cancer survivors should receive survivorship support to navigate online spaces effectively. Strengthening these multilevel supports can enhance the sustainability and psychosocial benefits of online cancer communities for YAs.