Misinformation about mental health has been identified across various online spaces, including Facebook, Reddit, and, prominently, TikTok. In 2021, the U.S. Surgeon General even named health misinformation as a public health crisis, and for good reason. Mental health misinformation can:
Increase stigma and perpetuate prejudice, discrimination, and rejection from others
Pathologize normal experiences
Misattribute symptoms of one disorder to another (i.e., some diagnoses are more “popular” than others on social media)
All of which ultimately thwarts self-understanding, intensifies widespread problems of isolation and loneliness, and interferes with getting the appropriate treatment.
The frightening truth is that the public is regularly exposed to, believes, and acts on information that is inaccurate, misleading, and/or outright harmful. However, the average person still faces real barriers to accessing accurate information from trustworthy sources. So, as you and others in your life voyage out into the murky waters of online self-education, consider what recent research has revealed about online mental health misinformation, what increases the risk of being misled, and how we can protect ourselves against misleading or outright false pieces of information.
What’s Out There
Starvaggi and colleagues (2024) reviewed, summarized, and offered reflections on the state of mental health misinformation on social media. The article presents an overall concerning picture of mental health misinformation on TikTok (i.e., 31 percent deemed “scientifically inaccurate” and 14 percent “potentially damaging”). Nonetheless, it seems that certain topics are more subject to misinformation than others.
One study found misinformation to be particularly common around the subjects of suicide, personality disorders, psychosis, and treatment (Hudon et al., 2025), whereas another report found that trauma-related and neurodevelopmental disorders (e.g., ADHD and ASD) featured the highest rates of misinformation. By the numbers, this broke down to only 33 percent of trauma-related content being considered “scientifically accurate.” For neurodevelopmental disorders, the misinformation problems are varied.
For example, videos about ADHD (Yeung et al., 2022):
More scientifically accurate, but highest rate of “potentially damaging” content (18 percent)
Still, 52 percent of videos had “information lacking scientific evidence” (e.g., signs and symptoms of ADHD)
Videos about ASD (Aragon et al., 2023):
Of the videos reviewed, 27 percent were accurate, 41 percent were inaccurate, and 32 percent were potentially misleading (e.g., overgeneralizing characteristics of individuals with ASD)
Vulnerability to Misinformation
Recent studies have found that greater exposure to misinformation and lower health literacy are two prominent risk factors for agreeing with mental health misinformation in online communities (Bizzotto et al., 2023; Hoffner et al., 2026). In other words, the more content that is out there and the more we see it (exposure), the more vulnerable we are to believing information that is inaccurate, misleading, and/or dangerous. Moreover, the less we know about mental health from reliable, accurate sources (health literacy), the greater our chances of being led astray.
Interestingly, self-efficacy appears to be another key variable in how people engage with misinformation online in that those with higher self-efficacy indicated greater intentions to respond to, rather than ignore, posts with misinformation. Meaning, people who believe they can have an active influence on others and the world around them act in ways that suggest a level of self-worth and empowerment to help protect others by preserving factual integrity.
In Safety We Trust (and Learn)
So long as there are real barriers to accessing quality mental health services and information, people in need will naturally, and resourcefully, find alternative means of care and education. The noticeable rise in young people turning to TikTok for mental health information during the COVID-19 pandemic reflected this quite well (Lee, 2022).
Furthermore, the issues of mistrust in the mental healthcare system are well-documented both off- and online. Individual and communal experiences of invalidation, underdiagnosis, and generally low-quality care reasonably motivate hurt, scared, and frustrated individuals to seek and find the validation they desire from peers in online communities. The real and important sense of safety and belongingness is one proposed link between exposure to and agreement with mental health misinformation. In contrast to the safety and trust derived in these communities, one study found that only half of the 37 percent of participants who were influenced by advice they encountered in online communities consulted a professional about the changes they were making in the management of their illness (Tan et al., 2021).
Social Media Essential Reads
Although rampant online, health misinformation isn’t new, and it isn’t going anywhere. While regulations and fact-checking mechanisms are useful safeguards, they do not address some of the most important underlying drivers for how misinformation gains traction in our minds and mental health decisions. When professionals do not satisfactorily respond to needs for empathy, validation, and trust, those professionals will, at best, sound like an adult in the Charlie Brown universe to a person seeking not only information, but also care and compassion.
Therefore, it is incumbent upon providers to address issues of trust in their interactions with patients and the public. However, as Jon Allen (2022) goes through great lengths to unpack, conceptualize, and offer some guidance on in his book, Trusting in Psychotherapy, becoming trustworthy as a mental health professional remains a major gap in formal training and education.
Yet, blaming individuals or even groups of professionals also likely misses the bigger picture. Healthcare professionals across settings are facing growing pressure to conform their approach to treatment according to the priorities of insurance companies and private equity groups, which privilege easily quantifiable, replicable, and financially conservative modes of treatment (Cotton, 2025; Michaels, 2025). In and of itself, this may not sound like a problem, but when these treatment models insidiously (and inaccurately) determine that the time and training required to genuinely establish trust with people who both need and fear mental health treatment the most is inefficient and too complex to quantify, then it is no wonder that people turn to the approachable voices online.