Abstract
The article uses the concept of online self-disclosure and examines whether TikTok videos reveal information similar to what is reported in existing research on social media within this field. In addition, the study aims to identify the creators’ motivations and the meanings they attribute to disclosing cancer and asks whether this disclosure challenges or supports the concept of a positive culture defined within the online self-disclosure framework. While similar research typically focuses on posts and is thus limited, this study combines conventional content analysis of 862 videos with in-depth interviews, offering a more nuanced understanding of the users’ lived experience. The findings highlight that TikTok is a platform for negative (e.g., fears, anxiety) and positive self-disclosure (e.g., joyful life events, self-acceptance). What is crucial, though, is that users feel exceedingly competent in educating others. The research debunks positive culture as an overly general category and undermines the traditional understanding of online self-disclosure. It reveals that seeking support is less important for patients than informing and warning others. By focusing on the educational meanings attributed to self-disclosure, this study enriches the body of research on cancer-related content shared on TikTok, including studies on low-quality cancer-related information.
Introduction
TikTok is a social media platform with millions of users and billions of downloads (Iqbal, 2023). It allows sharing of user-generated short videos between 15 s to 3 min—user interactions, shares, hashtags, and so on and determine content popularity. The users are primarily young, between 18 and 34 years old, and as of January 2024, only around 3% are 55 or older (Global TikTok User Age and Gender Distribution 2024, n.d.). This younger demographic makes the platform one of the most rapidly rising online enterprises, still scarcely researched, considering its relative novelty.
TikTok is recognized as a promotional tool and is frequently used for sharing health-related content with the potential to mobilize public health campaigns (Basch et al., 2021) although, the quality of health information varies based on its source (Kong et al., 2021; Song et al., 2021). There is existing literature on TikTok concerning COVID-19 (Basch et al., 2021), diabetes mellitus (Kong et al., 2021), mental health (Chadee & Evans, 2021), and chronic obstructive pulmonary disease (COPD) (Song et al., 2021). In addition, many studies tackle cancer-related health information quality (e.g., Abramson et al., 2023; Doyon et al., 2022; L.-S. Li et al., 2023; Morton et al., 2023; S. Yang et al., 2023). While it is a newer area of research, some studies address the sharing of cancer (and pre-cancer) narratives on other social media than TikTok and the internet more broadly (Finer, 2016; Foley et al., 2015; Hodson & O’Meara, 2023; Myrick et al., 2016; Stage et al., 2020; Wang & Wei, 2020; Wellman et al., 2023). All of these will be addressed in the Discussion section in this study context. Some new papers on TikTok are being published (Wellman et al., 2023); however, despite the aforementioned research on other platforms, the reasons people share their cancer-related stories on TikTok are still underexplored.
The present research addresses this lack and focuses on TikTok videos by users with cancer and their experiences. The study uses the theoretical framework of online self-disclosure and social media as a “positive culture,” the latter taken out and addressed from the perspective of online self-disclosure. Although we develop this particular field by addressing its research gaps, we do not resign from discussing research that is relevant from the perspective of this study but relates to other areas, such as cancer communication on social media in general. Although some of the studies we reference do not use online self-disclosure, these are essential in describing patients’ use of social media for information-seeking, helping other patients, and solving practical tasks (Stage et al., 2024); topics relevant also to self-disclosure as discussed in the concluding section of the article. However, it must be stressed that many of the papers we indicate reflect on social media in general (or specifically on Facebook, Instagram, blogs, etc.), and the focus of this study is specifically on TikTok taken up to enrich the self-disclosure (and positive culture) reflection. This is achieved by challenging the conventional notion of online self-disclosure, as patients prioritize informing and warning others over seeking support and emphasize the educational significance of posting videos.
Self-disclosure, being not only an online notion but a broader category of behavior (Jourard, 1971), means sharing personal information about oneself with others. It may develop and strengthen interpersonal relationships as people share personal beliefs, life facts, and emotions. Pennebaker and Seagal (1999) showed that telling one’s story and sharing traumatic experiences and negative feelings with others could lead to “emotional relief.” This relief in sharing is now apparent in online spaces, and researchers are paying more attention to the online self-disclosure phenomenon, that is, users providing personal, private information online. Moreover, online self-disclosure is reportedly faster and more in-depth as more information is disclosed (for different views on this, see Nguyen et al., 2012).
Research on online self-disclosure often presents it as a form of self-therapy (Basch et al., 2022; Karsay et al., 2019; Luo & Hancock, 2020; D. Yang et al., 2019), which is also true for studies on cancer patients (Andalibi et al., 2017; Dolev-Cohen & Barak, 2013; Malloch & Taylor, 2019; Stage, 2019; D. Yang et al., 2019). They share their everyday life but also remember moments from the past, reporting fears, anxiety, pain, weaknesses, and suffering associated with the treatment and body changes (negative self-disclosure). Also, social media narratives relate to positive emotions, joyful life events, self-acceptance, or self-affirmation (positive self-disclosure). Hence, online users can witness “cognitive work,” entailing a re-evaluation of negative aspects of illness and framing them into positive aspects, which helps to cope with the disease and gain a sense of control over one’s life (Kaufman & Whitehead, 2018).
Studies have demonstrated that online self-disclosure can offer various advantages, including enhancing relationships, improving mental health, and providing social support (Karsay et al., 2019). Disclosing health-related issues has been noted to be more prevalent in online health support groups than in general discussions online (Yang et al., 2017). This suggests that individuals are more willing to disclose personal health information in environments where they perceive more understanding, empathy, and support from others. This is important from the perspective of the present research because TikTok, as a social media platform, cannot be considered a typical support group often studied in the context of self-disclosure. Showing how people with cancer function on TikTok adds to the notion of self-disclosure, especially in showcasing its complexity.
This complexity was already demonstrated by Pluta (2022), who studied women with breast cancer using Instagram to share their experiences. They are not anonymous, exposing both photos and personal information, and they also refer to the people supporting them on their way to recovery (e.g., relatives, hospital staff, etc.). Hence, traditionally understood online self-disclosure cannot alone explain this sharing, where anonymity and lack of offline support are often emphasized (Barak et al., 2008; Rains, 2014). Rimé’s (2009) concept of social-emotional sharing explains this discrepancy (Pluta, 2022). According to this researcher, sharing both positive and negative emotions makes these emotions meaningful, and looking for others and bonding with them in times of stress or uncertainty is extremely important. Disclosing information about loved ones and friends creates and strengthens the woman’s so-called “attachment system.” This study adds to all this complexity even more and unravels it. It does so especially because it has been previously found that sharing personal information on social media can positively impact an individual’s well-being by enhancing their self-image, reducing symptoms of depression, and gaining a sense of autonomy, recognition, and belongingness (Rothschild & Aharony, 2022). We highlight another source of empowerment for individuals navigating health challenges and disclosing them online, which is a sense of educational influence.
What is more, existing research connects online self-disclosure with the notion of social media “positive culture,” defined as showing oneself without any undesirable traits as being more advisable and accepted than content with a negative connotation (Hollenbaugh, 2021; Lee-Won et al., 2014). We use positive culture as part of the reflection on self-disclosure, and as another way, this study enriches the self-disclosure field. For example, Zhang et al. (2022) explored how disclosing positive versus. negative Facebook posts affects users’ willingness to interact with the discloser. “Positive self-disclosers” are perceived as more socially attractive than “negative” disclosers and may receive more public reactions. Similarly, Holtz and Kanthawala (2020) sampled 423 Instagram posts to understand individuals’ type 1 diabetes self-disclosure better. Overall, they noticed a greater prevalence of posts expressing positive sentiment than those conveying negative sentiment, suggesting the potential existence of a connection between self-disclosure and the manifestation of positive sentiment.
Online self-disclosure analyses typically revolve around posts or messages (e.g., Andalibi et al., 2017; Barak & Gluck-Ofri, 2007; Basch & MacLean, 2019; Kaal et al., 2018; Stüwe & Wegner, 2020; Tichon & Shapiro, 2003). Thus, the resulting literature is limited in concluding the users’ experiences. Furthermore, one cannot emphasize the positive effects of online self-disclosure only with content analysis, whether automated or manual. Thus, there is a need for more in-depth studies that go beyond pure content analysis to a more nuanced understanding of the lived experience of patients. This does not imply that studies using deep qualitative methodologies are absent in health communication, particularly regarding cancer patients’ narratives. One notable example is the Vital Media paper (Stage et al., 2020) along with other works mentioned in the Discussion. Despite these studies, we believe that more is needed, especially in self-disclosure research, and this is another way the present paper adds to this field. Contrary to many previous studies, this article connects content analysis with a more in-depth look at the users, going beyond a simple description of the content and utilizing in-depth interviews (IDIs).
Overall, we ask the following questions:
Combining CCA with IDIs was crucial for the research design. Stage 1 of the research involved analyzing the content using hashtags to identify and sample the most relevant videos (e.g., Fowler et al., 2022). Stage 2 (guided by stage 1), that is, structured online IDIs, expanded and added to content analysis. We discuss the ethics for each stage separately, here summarizing that the study received ethical approval from the Institute of Social Communication and Media Institutional Review Board (IRB) at the Kazimierz Wielki University in Bydgoszcz, Poland.
Because the research was carried out in two stages with different methods, in the next section, we combine the Methods and Results to present these separately for each stage. This approach is also optimal because the second stage directly results from the first. Finally, the article closes with a Discussion, focusing on the research’s broader meaning and its limitations.
Methods and Results
Methods: Stage 1—CCA
As the study is Polish-based, TikTok’s internal search engine and the hashtag: “nowotwór” (“cancer”; 24.1 million views as of December 13, 2023) was used along with three related hashtags: “nowotwor” (without the Polish “ó”; 12.7 million), “nowotwórzłośliwy” (“life-threatening cancer”; 2.6 million), “nowotwory” (plural for “nowotwór”; 944.8 thousand views). We identified these as the most relevant search terms based on initial scoping searches on TikTok. The searches validated the hashtags used in cancer self-disclosure videos with other terms producing only noise (e.g., “rak,” a colloquial for cancer in Poland, but that also means zodiac sign). Also, a new web browser and newly created profile with no identifiable search history or cookies enabled were used at every search for relevant content.
The videos tagged with the identified hashtags produced the highest number of views. The first 300 videos for each search were considered, and these videos directed us to the most relevant profiles (of users with different types of cancer and different stages of the disease). For each relevant profile, several characteristics were recorded: the number of followers, likes, and videos posted. Each of these was given an appropriate rank (from 1 to 300). The average of these three ranks aided in developing a list starting with the most popular and most content-rich profiles at the top. The list served to select the profiles and respondents (described in a later section). We chose the first four profiles from the list and analyzed all the videos represented. We considered cross-validating the findings with a smaller sample of TikTok data from less-popular accounts. However, we dropped this idea, as the content-rich profiles met our RQs and due to the lack of representativeness of the whole study (see the Discussion).
A data set of 1,054 videos was collected in October 2023. All retrieved videos’ URLs and metadata (e.g., username) were scrapped and saved using a TikTok Scrapper tool Apify. This tool was used only to automatically record metadata of the videos previously sampled manually. We also double-checked several videos for reliability. Of the 1,054 videos, we excluded those unrelated to cancer, as some videos could not relate to cancer at all. Also, cancer content that did not focus on the creator was not included, e.g., documentaries about cancer, news about famous people having cancer, and so on. The final sample consisted of 862 videos that featured cancer self-disclosure as the primary theme. The sampling procedure was based on similar studies on TikTok videos already mentioned in this article (e.g., Andalibi et al., 2017; Basch & MacLean, 2019; Kaal et al., 2018).
The study’s codebook was developed using a CCA approach as described by Hsieh and Shannon (2005) and in line with other content-based TikTok research (Boatman et al., 2022; Y. Li et al., 2021). The CCA is an inductive approach that assumes that categories “flow from the data” while researchers avoid preconceived notions. First, we reviewed the videos repeatedly (including making transcripts and a short note on the content of each video) to obtain a sense of the whole. Next, research team members independently generated codes/themes, compared the degree of overlap between the lists, and developed the final coding subcategories. Then, broader categories through discussions were introduced (see Table 1 in the Results). According to CCA, categories can represent explicit or inferred communication; thus, all subcategories and categories were clearly defined (see Table 1). Unlike many other studies, we did not focus on quantifying user engagement (e.g., Y. Li et al., 2021). Hence, the number of views, likes, comments, and shares is not reported. The same is true of information concerning the video format and length, subtitles, caption, and music.
Content’s Categories and Subcategories and Their Definitions.
Source. Own study.
Two coders (research assistants) independently assigned 862 videos to existing categories and subcategories, and they were instructed on the multi-labeled approach to coding. In addition, they were instructed that if they believed the content did not fit the preconceived categories, they had the freedom to indicate new ones to increase the validity of the coding. The coders were trained in two sessions of approximately 2 hours. During training, the theoretical assumptions of the research were explained. They were shown some videos (not included in the primary sample) and asked to assign them to appropriate categories. Also, any discrepancies were discussed and resolved. In the actual study, the sampled videos were coded by each coder, and intercoder reliability was high, with Krippendorff’s Alpha (α) values ranging from 0.88 to 0.96 (see Table 1). No new categories were indicated.
We did not violate the users’ privacy or cause them any harm. All users on TikTok are aware that their videos are accessible to anyone. Despite this, we believe the creators should be protected, as the videos concern issues considered private, intimate, and particularly sensitive (Elgesem, 2015). Hence, in our analyses, the likelihood of identification was eliminated. We aggregated and categorized the videos without describing any in detail, nor did we provide any video transcripts or images. Although we have given here hashtags used to reach the analyzed material, it would be highly unlikely to identify the analyzed profiles due to the mentioned aggregation and TikTok and the dynamically changing TikTok content and algorithms. In the following subsection, when reporting about the videos, categories, and/or subcategories are indicated (e.g., C01, SC01).
Results: Stage 1—CCA
Content analysis showed that the self-disclosure on TikTok is more positive (C02) than negative (C01) (Table 1). It is not that negative content is absent because videos inform about weaknesses and fears. Users talk about or present demanding situations, for example, showing themselves without hair due to chemotherapy or while waiting for medical examinations (SC01; SC02; SC03). They indicate how harsh the treatment can be and how many doubts they have; these videos can be retrospective (SC02; SC03). However, there is less such material than positive self-disclosure, that is, sharing “successes” related to, for example, treatment and showing that despite the disease, patients can lead an exciting life, have fun, and enjoy themselves (SC04; SC07). Furthermore, those videos convey that you should not give up (SC05) and accept an illness (SC04; SC06).
The positive tone of TikTok is also because even if the videos are negative, the message is often “softened.” This softening is because of so-called trends, which will be mentioned later. Another way is to combine negative content with positive content. One example is flashbacks, where a person shows how they looked when ill and treated, such as without hair, eyebrows, or eyelashes. These images are then juxtaposed with more current imagery, that is, with hair, makeup, and generally a person full of energy. Disclosing the user as a patient simultaneously attempts to turn an illness into something positive (SC02; SC03; SC04; SC06). Positive self-disclosure on TikTok is mostly educational (SC08). These videos are about preventing cancer, but the users also want to debunk myths and popular beliefs about this illness and, by doing so, normalize it. This will be elaborated on when discussing the interviews, as educating others was essential to the respondents’ experience.
Category 3 (C03), that is, disclosing information about other people (and support they provide), is not that numerous, which does not mean there are no such videos. Some feature family or friends, usually in funny, positive situations, with users expressing their thanks for the support they receive (SC09). Interestingly, if the videos referred to other people, this was mostly a reaction to negative or positive user comments (SC10). We described this in more detail in the next part of the article because the respondents also referred to this type of content.
Methods: Stage 2—In-Depth Interviews
The IDIs were used to achieve the previously mentioned more-nuanced understanding of the lived experience of patients. The previously indicated profile list was used to sample the respondents. In addition, we wanted to interview the users whose profiles were analyzed along with additional creators. Moving from the top of the list (the most content-rich accounts), each user was contacted by direct messages to arrange an interview. If a respondent did not agree to an interview or did not respond for several days, the procedure was repeated for the following user on the list. We reached and interviewed 13 people, including the creators of four analyzed profiles. The study reached sample saturation (e.g., Saunders et al., 2018) as the responses indicated similar themes and motives. We iteratively analyzed the data to ensure saturation and documented the entire process. Considering the qualitative nature of the research, we chose not to multiply the number of interviewees, especially since some researchers claim there are no general rules for sample size in qualitative research (Patton, 1990). The judgment on the sample size tends to be implicit (van Rijnsoever, 2017).
We provided participants with detailed study information before they agreed to participate in the interviews. This information outlined the purpose of the research, that participants’ involvement was utterly voluntary, provided sample questions and likely discussion directions, and explained the data collection methodologies. All respondents gave verbal consent and were informed they could choose not to respond to questions or resign from participating at any time.
Bearing in mind ethics and the sensitive nature of the study, we do not provide respondents’ details (Zimmer & Kinder-Kurlanda, 2017). The dynamically changing content on TikTok, algorithmic processes, a large number of profiles, and a relatively small sample in this study make it virtually impossible to uncover the identity of interviewees. However, to minimize this risk even further, we anonymized the respondents’ statements by not including or changing parts that could help identify them. Also, the participants were assigned numbers from I01 to I13, and in the Results section—when reporting about their statements—we additionally indicated the CCA category associated with each point of view. For example, I01SC01 means Respondent 01, and their statement was categorized as relevant for Subcategory 01. In the case of the statement being too general or not directly connected to the subcategories; the respondent’s number (e.g., I13) was provided. As these were structured IDIs, most of the questions were relevant to the categories identified in Table 1 and in line with the objectives of this study. Nevertheless, the interviewees were free to expand the predefined threads and introduce new ones as they saw fit with our minimal interference. We report these findings in the next section.
Results: Stage 2—In-Depth Interviews
The use of CCA and focusing on “positive content,” “negative content,” and “content about other people” in the first stage were quite categorical. As previously mentioned, the interviews “deepened” our understanding of the self-disclosure processes and informed us about users’ meaning-making and emotional work. When asked why they maintain a profile in the first place, the respondents claimed they mostly want to educate others, which was in line with what was evident in the content analysis. Notably, the interviewees did not understand education only as prevention. They reported posting videos about the importance of regular medical examinations, how to spot the first symptoms of cancer, or how crucial it is to lead a healthy life. However, this is only one of the educational aspects because they also want to “answer other people’s questions” (I03SC08). These may concern symptoms, treatment, and side effects or complications that may persist for an extended period. Thus, online self-disclosure has a very strong educational component.
In addition, all respondents indicated they wanted to challenge the popular beliefs surrounding cancer. For example, you cannot recover, and everyone is doomed to chemotherapy. These responses were not about strong warrior-like cancer patients (a topic indicated in some other studies discussed in the concluding section) but instead, these were framed as an essential educational theme and the desire to normalize the disease, that is, not treating it as something “extraordinary” (I04SC08), “foreign” (I12SC08), or “abnormal” (I13SC08). According to the interviewees, it is necessary to make people aware that cancer can happen to anyone. Still, it may not always mean “the end of the world” (I07SC04), especially since “early detection gives a good chance for full recovery” (I07SC08), and it is important to “emphasize it” (I01SC08).
All respondents claimed that they are perfectly capable of informing and educating others. This feeling of competence is an interesting theme that was not obvious during the content analysis as it was impossible to say whether the users feel competent despite posting educational content. The sense of competence—not given directly in the analyzed videos—results from being sick. Interviewees think that due to their experiences, they “have a certain mission to carry out” (I03SC08), that is, warning and informing others. According to the respondents, this feeling of being “the right person in the right place” (I03SC08) is particularly suitable for TikTok: It can be said that on TikTok—and in general on the internet—everyone is an expert in something. There are people specializing in just being fun and in being stupid . . . well, there is a lot of nonsense out there . . . But there are also doctors or others like that. And I’m just a cancer person, because I’m going through it myself and I just know what it looks like from the inside (I02SC10).
Importantly, respondents said little about using TikTok to share their fears and anxieties and to feel better despite the CCA showing many videos of traumatic experiences and difficult moments. It is not that they did not mention this at all because each respondent admitted to “sometimes just complaining” (I03SC01) and using TikTok to express suffering, weaknesses, and so on, as seen during the CCA stage. Nevertheless, sharing traumatic experiences was always framed in terms of a “call to act” (I02SC04). Showing weakness becomes an impulse to post educational or positive content, for example, to encourage others not to give up, or to show that having cancer, one can “lead a normal life, fulfill oneself in various life roles” (I06SC07), or that one “can’t give up” (I02SC05). However, as stated above, it is not about fighting cancer and being strong above anything else. Again, it is more about educating because “posting can be comforting, and at the same time, it’s about showing (others) that even with cancer, you can be very satisfied with life” (I09SC07). This theme was also notable during the content analysis stage, where positive self-disclosure was about, among others, showing oneself as a patient but also a family person, a worker, a traveler, and so on.
CCA evidenced that some of the profiles show subsequent phases of the diseases and are a step-by-step report on therapies, medication, treatment, and the results these bring. When asked whether their profile could be treated as a kind of diary, the respondents claimed that it was not the case but more about “finding a meaning” (I03SC04), “for oneself and for others” (I06SC08). “Seeking meaning” (I11SC05), however, is not the same as seeking support; “that’s not what this is about” (I02), but precisely about being able to educate others. It is about feeling “you’re doing something important, turning your experience into something good” (I08SC08).
Some respondents mentioned differences between TikTok and other social media, acknowledging that the so-called trends influence posting on TikTok. It is one of the most important parts of the TikTok experience, often binding communities that focus on current trends (Jaudon, 2022). These include hashtags, music tracks, filters or effects, and even a particular dance. Users create and popularize new trends with influencers encouraging followers to post their interpretations of a given trend. Trends are also followed and/or created by people with cancer: I always try to keep track of what is happening at the moment, what trends are currently popular. I try to use these, and this could be a popular song or filters. I can’t say that everything I post is related to trends because it’s not, but when I talk about cancer, I try to use trends, and I adapt these to cancer issues (I07).
Our respondents associated trends with a positive culture, often indicating popular trends that usually refer to great looks, family happiness, humor, and present successes. Those themes accompany music, colors, and movement (e.g., dance) (Jaudon, 2022). In this sense, the cancer content is indeed “lighter.” However, it is not that the users deliberately want to adjust to positive culture to “distort” or “hide” the harsh reality and present themselves in a better light. This does not align with the notion of positive culture as a need to present oneself without undesirable traits. The respondents claimed that maintaining a profile means genuinely presenting themselves truthfully and discussing some “harsh topics” (I10SC03). They reshaped dominant trends experiences on TikTok while noting that trends are great tools for reaching other people with cancer-related content because these correlate with TikTok algorithms. Even someone not interested in cancer videos can occasionally come across this content when it is within a given trend. According to interviewees, this is why TikTok differs from other social media, including Instagram. Algorithms also rule Instagram, but “if you don’t search for a given content yourself, for example, by a hashtag, the chances of reaching posts you did not search for in the first place are quite low” (I02).
In the case of this study, the positive culture means the dominance of content from the second category (e.g., the aforementioned flashbacks). However, not a single respondent said they wanted to increase the number of followers using this positive tone. The interviewees did not believe that due to dealing with difficult themes, their videos are inherently less popular or less visible, hence needing a “positive boost” (I05SC04). It was the opposite, as presenting oneself as a cancer patient is a reason for pride, and they are not ashamed of their illness. Hence, positive culture is more about using trends rather than deliberately using cancer videos to gain followers. The interviewees focus on themselves without considering how the cancer content influences their popularity, which is a specific paradox as TikTok’s algorithms are designed to maximize user engagement. They favor content that quickly gains popularity, which often means following the latest trends. The use of trends conflicts with the declared lack of interest in gaining followers, which could be explained by respondents’ educational motivations. They claimed that educating is not the same as gaining “empty followers” (I02), often random and uninterested in the topic. The interviewees’ reflections on gaining popularity, trends, algorithms, and education were therefore nuanced, and pressure to follow trends was a pressure to post to the audience, not for popularity measured by numbers but by educational influence.
Despite the lack of interest in gaining followers, one of the respondents mentioned negative comments from other users accusing her of “using illness to be more recognizable” (I06SC10). When asked how she feels about it, she said that she usually turns it into a joke, does not care, ignores it, and “thinks it’s not about self-promotion, because if anything, she’d rather gain followers in a different way than through using cancer” (I06SC10). At the same time, she emphasized that many people support her writing nice comments. Other respondents confirmed that positive feedback is common, and they like responding to it, especially answering questions about the disease. This sentiment confirms the previously discussed educational tone of the profiles.
Respondents emphasized that compared to other social media, TikTok is particularly suitable for informing about the disease, sharing experiences, and educating others. Those who use other platforms usually stated that they “feel that only through TikTok it is possible to reach someone” (I08SC08), and creating videos and answering comments gives them great satisfaction. Moreover, all interviewees stressed that they believe their audiences expect them to be more educational rather than posting only “sad content without any deeper, educational meaning” (I08SC08).
Discussion
When considering the first research question (RQ1), this study shows that TikTok is a tool for negative and positive self-disclosure and showing others who support in illness. However, it also indicates more positive content and the creators’ strong attachment to TikTok’s educational possibilities. At the same time, creating videos is strongly influenced by the algorithmic rules of the platform, most of all trends. This alone means that videos, although dealing with the difficult theme of cancer, can be perceived as “lighter” (more positive). Although we did not aim to thoroughly compare different platform cultures (e.g., Instagram vs. TikTok), findings suggest that respondents see the differences in social media for specific uses. They feel they want to make the best out of the algorithmic-driven recommendation mechanism of TikTok (the trends) to educate others. This became the only major issue related to how the architecture of a given platform may influence self-disclosure. This also aligns with studies reporting various online spaces (forums, support groups, social media, etc.) differing in the scope and degree of self-disclosure (e.g., Barak & Gluck-Ofri, 2007).
For example, Foley et al. (2015) conducted a qualitative content analysis of YouTube’s highest-rated breast cancer narratives, uncovering prevalent themes. The primary topic identified through transcript analysis was “the experience of fighting cancer,” with many patients expressing a desire to participate in peer support and survivorship programs. Similarly, Myrick et al. (2016) analyzed Twitter content over 2 years, examining the role of health-related social support and the impact of emotional expression on online social support in tweets about cancer. In a different approach, Wang and Wei (2020) introduced a deep learning model and social network analysis to explore the emotions present in a vast collection of cancer-related tweets. The results show that joy is the most frequently expressed emotion, followed by sadness and fear, with lower expressions of anger and hope. All this confirms Hodson and O’Meara’s (2023) claim regarding early-onset cancer patients, survivors, and caregivers using various social media platforms. They noted that these individuals tailor their social media posts to fit both the technical affordances of each platform and the anticipated audience on these platforms. They concluded that, especially on platforms like Instagram, the focus is on seeking information, social support, and hope. There is currently little research on TikTok specifically. Still, a 2023 study by Wellman and colleagues (2023) focused on previvors—people genetically predisposed to hereditary breast and ovarian cancer but not yet diagnosed. This study examines how previvors seek information and community support online to reduce uncertainty. It reveals that many previvors see it as their duty to share their genuine experiences on social platforms like TikTok and Instagram to aid others and lessen their uncertainties.
The existing research on cancer communication/narratives on social media includes multiple ways cancer patients engage with different platforms. For example, the Vital Media paper mentioned previously (Stage et al., 2020) investigated the lived experiences of adolescent cancer patients related to using Facebook, Instagram, and blogs. The primary insight of this in-depth interviews-based paper is that social networks are highly significant for young individuals following a cancer diagnosis. Thus, these networks can be called “vital media” and “experienced . . . on three different levels” (Stage et al., 2020, p. 10). First, they act as tools generating affect and are used by patients to share, seek, or avoid content to align themselves with desired feelings, manage time, and cope with existential crises. Second, these platforms are seen as unpredictable and dynamic in impacting patients’ moods, physical states, and perceptions of time. Third, they are described as having a collective “affective condition” enabling specific types of positive communication.
This highly interesting portrayal of the experiences of young cancer patients—although presented from a different theoretical perspective than online self-disclosure—highlights the diversity of experiences and ways of using social media by people with cancer (Stage, 2017; Stage et al., 2024). Hence, it shows that patients may not only be focused on therapy, support, and emotional relief, as in many self-disclosure studies. Having said that, it is crucial to note that this study reflects specifically on TikTok and aims to enrich the online self-disclosure field. The prevailing issue was that the users wanted to educate others, with education being a theme visible in the CCA stage but fully prevalent in the interviews. We reference this to self-disclosure and positive culture frameworks and suggest that self-disclosure reflection would do well to notice that TikTok cancer communication might be focused on educational purposes, an issue we will elaborate on below.
Regarding RQ2, we might say that IDIs did reveal a more nuanced lived experience of cancer patients using TikTok. While CCA shows that the content is educational, the interviews confirm this and reveal meanings that patients attribute to their online activities. They believe they are especially predisposed to inform and educate others through self-disclosure, as they know what it means to have cancer. We have thus achieved our methodological aim of noticing that when studying cancer experiences, it is crucial to use not only “non-invasive” methodological tools for content analysis (as is happening in many self-disclosure studies) but also go deeper into these experiences by using more in-depth methods.
In the context of RQ1 and RQ2, the present research challenges the traditional understanding of online self-disclosure. Similarly to other studies, we confirmed disclosers are not anonymous and receive support from others (e.g., Pluta, 2022). Meanwhile, the traditional approach sees the lack of this support as the motivation for self-disclosure and anonymity as its necessary condition (Barak et al., 2008; Rains, 2014). In addition, we undermine another pillar of the traditional approach, clarifying that disclosers may not necessarily seek support (for a similar view, see Lee et al., 2013; Weber et al., 2004). When looking at the content analysis results alone, one cannot decide whether receiving support is vital for creators. However, the interviews suggest that it may not be as crucial as one could assume. It is not that users do not value and do not need support, but the primary reason for maintaining profiles is to educate others. This notion is the most important for our respondents. Educating others is the experience they seek, and it helps them find meaning in their illness and suffering. Relating this to other studies on communication/narratives of cancer patients on social media, we might say that the respondents use TikTok to construct cancer narratives as a kind of project that uses illness in a manner characterized by Stage (2017). In this case, it is a specific, illness-based, and algorithmic-based educational project.
Regarding RQ3, the IDIs also show that for online self-disclosure on TikTok, using positive culture as recognized within the self-disclosure perspective mentioned in the Introduction may not be viable. This idea is especially true if positive culture is defined as presenting oneself without any undesirable traits and using this to gain popularity and gather followers. The respondents claim not to care how they are perceived, as they believe there is no need to build one’s image by avoiding harsh themes. For them, being natural and creating content to educate others is more important. Therefore, if one looks for positive culture on cancer patient profiles on TikTok, it has to be understood as a positive self-disclosure, as it was defined, that is, using flashbacks, trends, music, colors, and so on. Hence, this research shows that social media cannot be generalized in the context of positive culture understood from the perspective of online self-disclosure research. For self-disclosure scholars, it is equally important to understand that social media are internally culturally diverse, and users represent different niches, offer various content for different audiences, and “presenting oneself without any undesirable traits” may not matter to all. This has already been confirmed in studies on general digital and social media’s affective and experiential ambivalences (e.g., Phillips & Milner, 2017).
As was mentioned previously, the existing self-disclosure studies on cancer-related content on social media focus on video analysis and fall short of incorporating a more comprehensive qualitative approach. This limitation is also apparent in studies outside of the self-disclosure framework, notably in works that evaluate the quality of cancer-related health information on TikTok (e.g., Abramson et al., 2023; Doyon et al., 2022; L.-S. Li et al., 2023; Morton et al., 2023; S. Yang et al., 2023). These studies use a variety of medical methodologies, such as DISCERN, The Patient Education Materials Assessment Tool (PEMAT), the Video Information and Quality Index (VIQI), and Global Quality Score (GQS), to assess information quality on skin, gastric, prostate, gynecologic, thyroid, and breast cancer, among others. The findings generally indicate that the information quality ranges from poor to moderate. In the context of such research, the present one is important, as it introduces another qualitative layer by employing IDIs to explore users’ actual meanings about their creative engagement without evaluating the accuracy of the information shared. Examining cancer content on TikTok and, at the same time, interviewing patients/creators deepens our understanding of how and why people use social media in general and shows that this usage has many, often unexpected, facets and can thus be very nuanced. In addition, as indicated above, this study undermines the traditional understanding of online self-disclosure, as seeking support is less important for patients than informing and warning others, thus educating them. This adds an interesting twist to research stressing TikTok’s low quality of cancer-related information. This study underscores the significance and impact of understanding the intentions and meanings patients assign to their online activities outside of pure medical evaluation of the quality of videos posted. At the same time, we recognize that TikTok’s algorithms are built to boost user engagement by promoting content that rapidly becomes popular, typically aligning with current trends. As previously mentioned, creators aim to increase the reach of their posts and adapt their content to the platform’s algorithmic preferences. This could create pressure to post content that aligns with trends, regardless of its substantive value. However, we do not consider how this is problematic for spreading cancer misinformation. While this is a potential limitation of our study, it falls outside of its primary focus.
There are other limitations as well, mainly resulting from the methods chosen. Using the two-stage process described could have resulted in approaching the second stage with an informed bias, that is, evidence supporting a given result rather than non-supportive could have been more likely found. However, steps have been taken to reduce this problem. Even though structured, the interviews introduced a lot of freedom for respondents, as was already mentioned. In addition, to allow for neutral, objective, and unbiased results during the interview stage, we have not introduced the results of stage 1 to the interviewers. During the training, they received only the basic information necessary to conduct the interviews. Besides, credibility results from the specific interpretative stance, that is, looking for negative cases. While analyzing the interviews, we have put special attention on not missing results that are not in line with the stage 1 findings and are not only looking for those confirming these.
This study considers the most popular way of creating content on TikTok, that is, posting videos. The platform offers other possibilities, such as, live streaming. Thus, the disease narratives may vary depending on the tools used, and we considered only the most popular feature of TikTok. In addition, the volume of the videos analyzed is not representative—the same as that of interviewees. Instead, we consider the videos and respondents as “cases” that align with the study’s qualitative character (Flick, 2007). However, our analysis of a relatively large video corpus compensated for these limitations.
Future research could introduce representativeness related to variables not proposed here. This study is intended to be general in scope, as we did not differentiate between cancer types, analyze users’ socio-demographic characteristics, or include any cultural dimensions (e.g., users from Poland vs. those living in other countries). If these research threads are pursued in the future, this could further contribute to understanding self-disclosure in social media.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Financed by the Polish Ministry of Science under “Regionalna Inicjatywa Doskonałości” program.
An Ethical Statement
This study was approved by The Institute of Social Communication and Media Science Board (no approval no.). All respondents provided informed consent before enrollment in the study.
