3 - (IOP 6) The Answer to the Mental Health Crisis: Examining Media Bias Regarding Psychedelic Treatments for Depression and PTSD Using Natural Language Processing

Clinical and research interest in psychedelic-assisted psychotherapy for posttraumatic stress disorder (PTSD) and major depressive disorder (MDD) has increased rapidly over the last few years. This surge in research has resulted in enthusiastic media coverage of psychedelics as a treatment for these disorders that is perceived as biased compared to coverage of established treatments. Thus far, the potential bias of psychedelic-related media has been largely speculative. This bias is important to study, however, as it may impact patient enrollment in clinical trials as well as patient expectations, attitudes, and compliance with treatment, all of which may affect outcomes.

To measure potential media bias, we conducted pre-registered analyses comparing media coverage of psychedelics to FDA-approved antidepressants for major depressive disorder (MDD) and posttraumatic stress disorder (PTSD). We used LexisNexis to access publications from 2017 (when the FDA granted breakthrough status for the first psychedelic treatment) to 2024 and natural language processing to analyze >7,000 publications’ sentiment and content. Binary logistic regressions compared the language in paragraphs related to psychedelics, antidepressants, both, or neither. We also conducted exploratory analyses examining the "top topics" most correlated with media related to psychedelics vs. antidepressants.

As expected, in binomial regressions, positive emotion and sentiment were more strongly associated with media coverage of psychedelics than antidepressants (β=.21; β=.67, both p<.001), and negative emotion and risk language were more strongly associated with antidepressants than psychedelics (β=.25; β=.23, both p<.001). Contrary to expectations, reward language was more strongly related to antidepressants (β=.19, p<.001), and negative sentiment was more strongly related to psychedelics (β=.17, p<.001). Additionally, exploratory analyses identified that top topics in psychedelic-related media focused on psychedelics as a treatment for mental illness, regulatory oversight, and clinical trials. In antidepressant-related media, withdrawal symptoms, mechanisms, and prescribed uses of antidepressant medications were identified. These topics suggest that the difference in language in media related to psychedelics versus traditional antidepressants may be due to an emphasis on potential benefits of psychedelics and potential risks or side effects of antidepressants.

Overall, results suggest that media coverage emphasizes the purported benefits of psychedelics versus the drawbacks of established medications for MDD and PTSD. This bias may impact patient expectations, compliance with treatment, and ultimately outcomes. This analysis is the first of its kind to quantify the sentiment of media publications related to psychedelic-assisted psychotherapy and may provide empirical support for the idea that “the hype has gotten ahead of the science,” especially given the FDA’s recent rejection of MDMA as a treatment for PTSD. Guidelines for media on best practices related to reporting on intervention research may pave the way for a more balanced representation of treatment options and their potential benefits and risks.