Exposure to Mental Health Content on TikTok: Associations with Mood Disturbance and Body Image among Young Adult Women ()
1. Introduction
1.1. Background
The current study examines the association between TikTok use and psychological well-being, including mood, self-esteem, body image concerns, and mental health distress. A central aim of the study is to investigate psychological distress in relation to both the amount of time spent on TikTok and exposure to mental health-related content on the platform.
Recently, increasing concern has emerged regarding a teen mental health crisis (Belfort et al., 2024). Evidence suggests that rates of depressive symptoms—and, more concerningly, suicidality—have increased among adolescents, particularly among females (Keyes et al., 2019). Initially, researchers believed that these trends primarily reflected reduced social contact and social isolation during the COVID-19 pandemic (Nwabueze et al., 2025). However, levels of psychological distress did not substantially decline after the pandemic subsided. One potential contributing factor is increasing use of social media, particularly among adolescent females (Twenge et al., 2022).
A substantial body of literature has examined the relationship between social media use and psychological well-being (Keles et al., 2020). Although findings have not always been consistent, research generally suggests a positive association between time spent on social media and symptoms of psychological distress, depressive mood, anxiety, and body image dissatisfaction (Thorisdottir et al., 2020). These effects may be particularly pronounced among girls and young women.
Although prior studies have examined associations between TikTok use, self-esteem, depressed and anxious mood, and body image concerns among college-age women, relatively limited data are available regarding the use of TikTok specifically for accessing mental health-related content and the types of conditions most frequently viewed or discussed.
1.2. TikTok and Mental Health
TikTok currently has an estimated 1.7 billion users worldwide, with adolescent and young adult women representing one of its largest demographic groups. Originally developed as a platform emphasizing creativity through music, dance, and short entertainment videos, TikTok has increasingly been implicated in discussions surrounding adolescent mental health problems (Bahnweg & Omar, 2023; Conte et al., 2025).
A major concern is TikTok’s “For You” algorithm, which uses current and past viewing patterns to serve increasingly similar content to users. As a result, an adolescent experiencing transient sadness or anxiety may repeatedly encounter videos describing symptoms of depression, anxiety disorders, eating disorders, or other psychiatric conditions (Costello et al., 2023). Similar processes may occur for conditions such as mild autism spectrum disorder or Attention-Deficit/Hyperactivity Disorder (Foster & Ellis, 2024).
More recently, attention has focused on the possibility of social contagion of psychiatric symptoms via social media platforms. Rather than being viewed solely as stigmatizing, some mental health conditions may become normalized, socially reinforced, or even presented in ways that enhance social status online (Foulkes & Underhill, 2025).
Foulkes and colleagues suggest that part of the apparent increase in adolescents’ mental health symptoms may reflect the positive or celebrity-like portrayal of psychiatric conditions on social media. Increasingly, mental health concerns portrayed on TikTok are not always framed as conditions requiring professional treatment. Instead, they may be presented as positive, unique personality traits or characteristics that make individuals appear distinctive (Haltigan et al., 2023).
Although earlier platforms such as Tumblr and Instagram contributed to the rise of self-diagnosis, TikTok appears to have amplified this trend because of its highly personalized algorithm and short-form video format.
1.3. Social Contagion and Psychiatric Symptoms
Despite the benefits of raising mental health awareness on social media, some psychologists have expressed concern that social media may portray conditions inaccurately and that psychiatric terminology is often used casually or imprecisely online, potentially trivializing the experiences of individuals with formally diagnosed conditions (Foulkes & Underhill, 2025). For example, diagnostic labels such as “bipolar” may be used loosely to describe normal mood changes.
One frequently cited example of possible social contagion involves Tourette-like symptoms with relatively sudden onset, often involving prominent upper-body tics (Frey et al., 2022). In classic Tourette syndrome, motor tics typically precede vocal tics, and fewer than 15% of individuals demonstrate severe vocal symptoms (Frey et al., 2022).
Another example involves Dissociative Identity Disorder, a relatively rare condition with an estimated prevalence of approximately 1%, which has gained increased visibility on TikTok. Haltigan et al. (2023) suggest that rare disorders may become especially popular online because they emphasize individuality and uniqueness.
MacMillan (2024) also identifies Attention-Deficit/Hyperactivity Disorder, Borderline Personality Disorder, and autism spectrum disorder as conditions frequently represented in TikTok-related content. MacMillan (2024) describes some of these online presentations as “performative sick role content” and argues that certain displays may resemble malingering or factitious disorder. Researchers have also expressed concern that social contagion processes may contribute to the spread of conversion-like symptoms and other functional neurological presentations.
1.4. Self-Diagnosis and Social Media
The increase in self-diagnosis of mental health problems has been cited as one factor contributing to the apparent rise in mental health conditions among adolescents and young adults. Self-diagnosis occurs when an individual concludes that they meet criteria for a psychiatric condition without undergoing a professional assessment (Rutter et al., 2023).
Armstrong et al. (2025) found that even among youth seeking professional mental health services, more than 70% had already diagnosed themselves prior to evaluation and reported that social media engagement contributed to their self-diagnosis.
Qualitative research has also found that many young people view self-diagnosis as acceptable when it provides a framework for understanding and legitimizing emotional experiences. At the same time, participants generally did not support the use of psychiatric labels as a means of avoiding personal responsibility (Armstrong et al., 2025).
As mentioned above, some researchers have suggested that self-diagnosis and online self-presentation as mentally ill may function as strategies for enhancing self-worth, obtaining social recognition, or improving short-term mood states (Haltigan et al., 2023).
1.5. Social Comparison, Mood, and Body Image
Social comparison has been identified as one of the key mechanisms linking social media use to depressive symptoms and body image dissatisfaction. Upward social comparison occurs when individuals compare themselves to peers whom they perceive as more attractive, successful, or socially desirable.
Social media platforms frequently present highly curated and idealized depictions of daily life, increasing the likelihood of upward social comparison. This process has been associated with envy, jealousy, irritability, anxiety, and depressive symptoms (Panchal et al., 2024). Research has also shown that individuals who compare their physical appearance to celebrities, influencers, or peers perceived as more attractive are more likely to experience body dissatisfaction and develop depressive symptoms (Charmaraman et al., 2021).
Another proposed mechanism is cognitive overload (Yan et al., 2024). Because social media users are exposed to large amounts of emotionally salient information while managing interpersonal interactions, excessive social media engagement may overwhelm cognitive and emotional resources. Cognitive overload theory suggests that this process may contribute to fatigue, stress, and dysphoric mood (Khalid et al., 2026; Yan et al., 2024).
Yan et al. (2024), for example, found that social media use among college students was positively associated with both cognitive overload and depressive symptoms, and that cognitive overload partially mediated the relationship between social media use and depression.
1.6. Current Study
The current study explored the level of engagement with TikTok, including the amount of time participants spent viewing mental health-related content, particularly material related to depression and anxiety.
A secondary purpose of the study was to examine the association between TikTok use and symptoms of depression and anxiety, self-esteem, and body image discrepancy. Although previous studies have identified associations between social media use and these psychological variables, findings have not always been consistent, and relatively few studies have focused specifically on TikTok.
Because social media use is often episodic and emotionally reactive, the current study used the briefer version (Curran et al., 1995) of the Profile of Mood States (McNair, 1992; Searight & Montone, 2017) to assess short-term aversive emotional states.
Overall, the study seeks to clarify the relationship between TikTok engagement, exposure to mental health-related content, social comparison processes, and psychological well-being among young adult women.
2. Methodology
2.1. Participants
Participants were recruited through Prolific, an online research platform used for behavioral science studies that matches participant profiles to studies for which they are qualified. The eligibility criteria specified that women must be currently enrolled in a higher education institution. Prolific reports a proprietary method for detecting fraudulent responses and ensuring data quality (Prolific, 2026).
The sample consisted of 151 college-aged women ranging in age from 18 to 27 years (M = 22.56, SD = 1.90). The sample was racially diverse. Participants identified as Black/African American (49.0%, n = 77), White (34.4%, n = 54), Asian (3.8%, n = 6), Multiracial (3.8%, n = 6), Hispanic/Latina (2.5%, n = 4), and Middle Eastern (2.5%, n = 4).
Most participants were undergraduate students (55.4%, n = 87). Additional participants reported enrolment in a master’s program (31.8%, n = 50), doctoral program (1.3%, n = 2), vocational or trade school (1.3%, n = 2), or community college (1.9%, n = 3).
Regarding mental health status, 31.8% (n = 50) reported having received a formal mental health diagnosis, 18.5% (n = 29) reported being self-diagnosed or suspecting that they had a mental health condition, and 45.9% (n = 72) reported no diagnosis.
2.2. Assessment Methods
2.2.1. TikTok Usage Questionnaire
Participants responded to the following questions regarding their TikTok use by selecting one of the provided response options.
1) Length of TikTok Use
Participants indicated how long they had been using TikTok. Response options ranged from less than 6 months to more than 5 years.
2) Average Daily Time Spent on TikTok
Participants estimated the average number of minutes per day they spent on TikTok. Response options ranged from less than 30 minutes to more than 300 minutes per day.
3) Frequency of Checking TikTok
Participants reported how frequently they checked TikTok. Response options ranged from multiple times per hour to less than once per week.
4) Exposure to Mental Health-Related Content
Participants estimated the percentage of TikTok content with which they interacted (e.g., liked, commented on, shared, or watched) that related to mental health or emotional topics. Response options ranged from 0% - 10% to 91% or greater.
Examples of mental health-related content included anxiety, depression, therapy, trauma, relationships, emotional struggles, coping mechanisms, mental health awareness, self-help, and emotional support.
5) Types of Mental Health Content Encountered
Participants identified the types of mental health or emotional content they most frequently encountered on TikTok by selecting all applicable options. Eleven categories were provided, including anxiety, depression, post-traumatic stress disorder (PTSD), and Attention-Deficit/Hyperactivity Disorder (ADHD)/neurodivergence.
6) Engagement With Mental Health Content
Participants described their level of engagement with mental health-related content on TikTok. Response options ranged from actively seeking out this content to rarely encountering it.
2.2.2. Rosenberg Self-Esteem Scale
The Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965) was used to assess global self-worth and self-acceptance. The RSES consists of 10 items rated on a 4-point Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree), with five items reverse scored. Although some researchers have proposed a two-factor structure separating positively and negatively worded items (Donnellan et al., 2016), the scale is most commonly treated as a unidimensional measure of global self-esteem.
The RSES has demonstrated strong psychometric properties across diverse populations. Internal consistency reliability coefficients typically range from .88 to .95 (Sinclair et al., 2010), and test-retest reliability over two-week intervals has ranged from .85 to .88 (Robins et al., 2001). Evidence supporting construct validity includes significant associations with related constructs such as depression (r = −.54) and life satisfaction (r = .47), as well as factorial invariance across gender, age, and ethnic groups (Schmitt & Allik, 2005).
2.2.3. Profile of Mood States-Short Form
The Profile of Mood States-Short Form (POMS-SF; Shacham, 1983) was used to assess transient mood states and psychological distress over a specified time interval. The measure consists of 37 adjectives rated on a 5-point Likert scale ranging from 0 (not at all) to 4 (extremely), indicating the extent to which each adjective described the respondent during the previous week.
The POMS-SF yields scores on six subscales: Tension-Anxiety, Anger-Hostility, Depression-Dejection, Vigor-Activity, Fatigue-Inertia, and Confusion-Bewilderment. In addition, a Total Mood Disturbance score can be calculated by summing the five negative mood subscales and subtracting the Vigor-Activity score. Example items include adjectives such as “tense,” “angry,” “unhappy,” “lively,” “worn out,” and “confused” (McNair et al., 1981).
The POMS-SF has demonstrated strong psychometric properties across both clinical and nonclinical populations. Curran et al. (1995), in a sample of approximately 600 participants, reported Cronbach’s alpha coefficients ranging from .76 to .95 across subscales, with most coefficients exceeding .90.
2.2.4. Stunkard Figure Rating Scale
The Stunkard Figure Rating Scale (FRS; Stunkard et al., 1983) was used to assess body image perception and body dissatisfaction. The FRS is a widely used pictorial measure consisting of nine silhouette figures that range from very thin to obese. The measure was developed to provide a rapid assessment of body image while minimizing literacy demands and has become one of the most frequently used instruments in body image research (Thompson & Gray, 1995).
Participants are asked to identify the figure that best represents their current body size and the figure that represents their ideal body size. Body dissatisfaction is calculated as the discrepancy between current and ideal body figure selections, with larger discrepancies reflecting greater body image dissatisfaction or disturbance (Gardner et al., 1999).
The FRS has demonstrated acceptable psychometric properties, including good test-retest reliability and convergent validity with other measures of body image dissatisfaction and eating pathology.
The FRS has demonstrated good test-retest reliability over a two-week period (r = .89).
2.3. Procedure
Participants were recruited through Prolific, an online participant recruitment platform. After accessing the study link, participants were presented with an informed consent form describing the purposes of the study, research procedures, potential risks and benefits, confidentiality protections, and the voluntary nature of participation.
Because the survey was administered online, implied consent procedures were used. Participants were informed that continuing with the survey indicated consent to participate, whereas closing the browser window or discontinuing the survey indicated a decision not to participate.
Prior to implementation, the study protocol was assessed and approved by the Lake Superior State University Institutional Review.
3. Results
The mean scores, standard deviations, and ranges for the POMS subscales, the Rosenberg Self-Esteem Scale, and the body image discrepancy scores are presented in Table 1.
Participants reported substantial TikTok engagement, with 83.5% indicating use multiple times per day or more frequently Tok usage reportedly ranged from 15 to 330 minutes (See Figure 1).
For these 151 young women, the average percentage of mental-health related content viewed was 39.27% with exposure ranging from 5% to 95.5%. The largest proportion of participants fell within the 41% - 60% exposure category (see Figure 2).
The reported frequencies of categories of mental health content viewed are presented in Figure 3; 30% - 40% reported viewing content addressing Anxiety. Depression, and ADHD/Neurodivergence with slightly below 30% reporting viewing content on PTSD. However, in terms of popularity, Relationship content was the most commonly viewed topic.
Table 1. Descriptive statistics for mood, self-esteem, and body image variables (N = 151).
Variable |
M |
SD |
Min |
Max |
POMS Total Mood Disturbance |
35.67 |
15.72 |
9 |
81 |
POMS Tension |
7.73 |
3.47 |
0 |
18 |
POMS Depression |
11.28 |
4.13 |
4 |
24 |
POMS Anger |
9.07 |
6.29 |
0 |
26 |
POMS Confusion |
6.66 |
4.30 |
0 |
20 |
POMS Vigor |
4.44 |
4.69 |
0 |
24 |
POMS Fatigue |
5.37 |
3.90 |
0 |
19 |
Body Image Discrepancy |
1.42 |
1.01 |
0 |
4 |
RSES Self-Esteem |
18.72 |
6.42 |
1 |
30 |
Note. POMS = Profile of Mood States; RSES = Rosenberg Self-Esteem Scale. M = mean; SD = standard deviation; Min = minimum observed score; Max = maximum observed score.
Higher POMS scores reflect greater mood disturbance except for Vigor.
To assess the strength of association between TikTok use and the Rosenberg Self-Esteem, Profile of Mood States—SF and Body Image Discrepancy, correlations were computed. Average daily TikTok use (in minutes) was significantly associated with greater exposure to mental health-related content, r = .323, p < .001. Exposure to mental health-related content was positively correlated with POMS Tension, r = .198, p = .015, and POMS Depression, r = .296, p < .001. However, exposure to mental health-related content was not significantly associated with Total Mood Disturbance, r = .015, p = .200, POMS Fatigue, r = .128, p = .118, or POMS Confusion, r = .088, p = .282.
Body image discrepancy was significantly negatively correlated with self-esteem, r = −.242, p = .003, indicating that greater body image dissatisfaction was associated with lower self-esteem. Rosenberg Self-esteem was negatively correlated with Anger, r = −.385, p < .001, Confusion, r = −.366, p < .001, Fatigue, r = −.221, p = .006, Vigor, r = −.251, p = .002, and Total Mood Disturbance, r = −.289, p < .001.
Body image discrepancy showed a marginally significant negative association with percentage of mental health-related content viewed (r = -.146, p = .075), suggesting a trend in which greater exposure to mental health content was associated with lower body image discrepancy. Body image discrepancy was not significantly correlated with POMS Depression (r = .012, p = .880) or Total Mood Disturbance (r = .089, p = .279). As noted above, body image discrepancy showed a significant negative correlation with self-esteem (r = −.242, p = .003), indicating that greater body image distress was associated with lower self-esteem.
A hierarchical multiple regression examined predictors of body image discrepancy. Age and ethnicity entered in Step 1 significantly predicted body image discrepancy, F (2, 146) = 3.96, p = .021. Weekly TikTok use entered in Step 2 did not significantly improve model fit. In Step 3, mental health-related content exposure significantly increased explained variance (ΔR2 = .04, p = .013). In the final model, greater exposure to mental health content significantly predicted lower body image discrepancy (β = −.22, p = .013). The hierarchical regression predicting Total Mood Disturbance was not statistically significant at any step. Self-Esteem: The regression model predicting self-esteem was not significant overall. Although mental health content exposure reached significance as an individual predictor (β = .18, p = .049), the overall model was not statistically significant and should be interpreted cautiously.
Figure 1. Frequency of overall TikTok use.
Figure 2. Percentage of TikTok mental health content viewed. Note. Participants reported viewing an average of 39.27% mental health-related content exposure.
The largest proportion of participants fell within the 41% - 60% exposure category.
Figure 3. Frequency of mental health related topics viewed.
4. Discussion
The findings of the current study provide indirect support for the potential role of social media in mental health contagion (Vidal et al., 2020). Given the substantial amount of time participants reported spending on TikTok, the platform likely exerts a meaningful influence on attitudes, emotions, and self-perceptions. When considered alongside the relatively high level of exposure to mental health-related content, these findings suggest that TikTok engagement may have important psychological effects. Consistent with previous research, greater exposure to mental health-related TikTok content was associated with higher levels of depressed mood and tension/anxiety symptoms (Aalbers et al. 2019).
Among the categories of mental health-related content reportedly accessed, relationship issues were the most common. Given the developmental stage of this sample, such concerns are not unexpected. Relationship formation and maintenance represent central developmental tasks during emerging adulthood, particularly among college-aged women (Arnett, 2000). Social media platforms may intensify these concerns by increasing opportunities for social comparison, emotional validation, and exposure to interpersonal conflict narratives.
The limited associations between body image distress and TikTok use may reflect the composition of the sample, as nearly half of the participants identified as African American. Although findings have not always been consistent (Grabe & Hyde, 2006), a body of research suggests that body image concerns are generally less prevalent among African American women (Lovejoy, 2001). For example, Roberts et al. (2006) found that disparities in body image dissatisfaction between White and African American women were most pronounced among women in their early twenties, an age group that closely resembles the present sample.
Frequent exposure to mental health-related material may encourage emerging adult women to interpret their own emotional experiences through a diagnostic framework (Feucht, 2025). However, alternatively, participants experiencing greater emotional distress may have been drawn to the Mental Health Content on TikTok. Although the present study cannot determine causality, the findings are consistent with concerns that repeated exposure to mental health content on social media may contribute to self-diagnosis and the normalization of psychiatric labeling among young adults (Foulkes & Underhill, 2025).
The pattern of findings suggests that the nature of the content viewed may be more psychologically important than the overall time spent on the platform. Specifically, overall TikTok usage was not significantly associated with self-esteem, whereas exposure to mental-health-related content was associated with higher levels of tension and depressive mood. These findings support broader social media research suggesting that content type and patterns of engagement may exert stronger psychological effects than the simple duration of use (Frey et al., 2022).
The current study adds to the growing body of research linking social media use to anxiety and depressive symptoms (Shensa et al., 2018). In addition, it supports concerns regarding the potential iatrogenic effects of mental health-related content on platforms such as TikTok (Giedinghagen, 2023). The algorithm used by TikTok has been described as creating a recursive online environment in which users are repeatedly exposed to similar forms of content, potentially reinforcing identities centered around psychiatric symptoms (Romann & Oeldorf-Hirsch, 2025).