For this exploratory content analysis, a seeded sample of TikTok videos with the hashtag #k2spice were retrospectively identified on June 29, 2021, using structured searches on the platform without any user login or personal search history enabled. The most relevant videos to the hashtag as determined by the TikTok search results algorithm were displayed first in order of decreasing relevance. The URLs for all the public videos returned in searches were saved, and raw video data files were downloaded using a custom script built in the Python programming language. Based on a seeded sample of 89 TikTok videos of 53 unique users/creators, we then snowball sampled for additional content from TikTok users who were coded as being associated with K2/Spice content among purported prison users. The 10 most recent posts from each of these associated TikTok user public profile pages were then collected for further analysis.

Manual annotation and content analysis of the TikTok videos were conducted by TJM, AC, and MCN. Coding was approached inductively due to the exploratory nature of the study, prior studies that have examined TikTok primarily using an inductive coding approach for TikTok videos and texts, and due to the lack of existing studies that have coded TikTok content related to synthetic drugs [14]. Authors generated a data set with binary classifications of whether the video discussed content related to adverse effects, selling, and consumption of K2/Spice. Authors also recorded user engagement (views, likes, comments, and shares), whether a post reflected sale of a K2/Spice product, if a post was associated with other user comments indicating intent to buy or sell K2/Spice, and whether the video appeared to be recorded in a prison-based setting. Three authors (TJM, AC, and MCN) coded all TikTok videos independently and achieved a high intercoder reliability (κ=0.95) for codes. For inconsistent results, authors reviewed that video’s content and metadata with other authors and conferred on the correct classification. Pearson correlation coefficient was calculated to assess the association between user engagement metrics and characteristics indicating intent to buy or sell K2/Spice as well as the medium on which K2/Spice was purportedly sold.

An ethics exemption was not sought for this study. All information collected during this study was available in the public domain, and the study did not involve any interaction with users. Any user indefinable information is removed from study results, and results are provided in the aggregate to ensure anonymity.

This observational infoveillance study analyzed 89 TikTok videos using the hashtag #k2spice and found that 40% of the content included discussing the use, solicitation to purchase or sell, or adverse effects of K2/Spice among prison populations. We also found a significant positive relationship between interactions with TikToks that included selling and buying activity among users. These exploratory results provide additional evidence that K2/Spice is a drug subject to abuse among the incarcerated population, and discussions about its use and sourcing are actively occurring on social media. Specifically, we found that the harmful effects of synthetic cannabinoid use are being recorded and shared on TikTok, and these TikToks are surfacing on K2/Spice-specific hashtag/keyword searches, potentially exposing other users to this content [24].

Our study specifically identified videos and comments that provided seller contact information and purported proof of product, indicating that content detailing where and how K2/Spice can be obtained in prison systems including through discrete packaging is being posted. We note that this appears to be a direct violation of TikTok drug content policies [25]. Several user-posted videos also demonstrated ways to evade drug screening procedures in prisons, such as placing K2/Spice on envelopes and greeting cards or putting it in a nasal spray, making it more difficult to detect the drug. This is consistent with previous findings noting the risks of its evasiveness and increasing use among this population due to being colorless, odorless, and highly potent [26]. Importantly, this open promotion of the sourcing of K2/Spice in state and federal correctional facilities may mean that screening procedures in prison facilities are being actively circumvented, introducing unique public health and substance abuse safety risks that require further study and response.

Our general findings align with several local news reports describing adverse psychological effects from K2/Spice use in state prisons, though the adverse effects of these synthetic drugs occur in all populations including those specifically involved in polysubstance abuse [27,28]. However, these consequences may be exacerbated in a population known to face social stigma, limited access to health care, unique psychological and mental health pressures, and risk of physical harm due to incarceration [5,26]. Additionally, continuing use of contraband substances while incarcerated may delay parole and the timely reintegration of individuals incarcerated for minor drug offenses because federal correctional facilities require the successful completion of rehabilitation programs [5].

Our results also suggest that higher user engagement on videos displaying K2-related adverse effects is associated with more comments intending to buy or sell K2/Spice. This may indicate that content specific to sourcing behavior may lead to higher interaction among TikTok users viewing behavioral-related content, warranting more targeted content moderation. This was observed despite TikTok guidelines stating that any content depicting or promoting drug consumption or solicitation will be removed, and violating accounts, when warranted, will be reported to legal authorities [25]. As TikTok is one of the newest and fastest growing social media platforms, it can be difficult to proactively identify content that violates user guidelines. However, regulation and the need to develop novel surveillance technologies is important as multiple studies have already identified unflagged calls for violence, harmful “internet challenges” (eg, Tide pod challenge), and antisemitism on the platform [12,29,30].

Synthetic cannabinoids are unregulated and often used in combination with other substances, which may increase the risk of adverse health effects [31]. In the presumed absence of rigorous prison screening standards for synthetic cannabinoids, specific harm may be brought on incarcerated populations who have recognized disparities in access to substance use treatment and psychosocial support [32]. Synthetic cannabinoids are increasingly easier to obtain and are deceptively described as cannabis cessation tools, emphasizing the importance of screening and peer-to-peer education [9]. The means by which synthetic cannabinoids such as K2/Spice are transported into prisons are also an issue of concern. Cards and letters are soaked with synthetic cannabinoids and serve as a substrate for substances such as K2/Spice, with varying concentrations creating “hot spots” that are unlikely to be known by incarcerated individuals [33].

There has been an increased pressure on social media companies to regulate the sale of illegal drugs occurring on their platforms as overdoses from opioid use, intentional and accidental self-injuries, and fentanyl poisonings continue to rise in the United States. Social media surveillance may represent a viable means to identify new trends and behaviors associated with this ongoing public health threat, particularly among traditionally hard to reach populations. The increasing popularity of synthetic cannabinoids has not been widely described in the literature, which has made it difficult to develop evidence-based interventions and policies to mitigate its harm. Results from this study can inform future research seeking to further characterize the use of social media platforms to promote both the use and trafficking of synthetic cannabinoids as well as lay the foundation for more targeted interventions in these uniquely vulnerable populations. Future studies should further validate exploratory results generated from this study, using additional hashtags and keywords, as well as data from other social media platforms, and augment these findings with other qualitative and quantitative research.

This study has certain limitations. Results are exploratory in nature, limited to a convenience sample of public content on TikTok, and may not be indicative of broader substance use patterns or themes in the general population. This study was also limited in its data collection and sample of data by the TikTok search algorithm, which displays videos based on their relevance to the keyword(s) searched at a particular point in time. Keywords associated with substance use are often monitored and removed by social media platforms, leading users to use pseudonyms or code words to evade content moderation. In fact, as of this writing, the term “k2spice” is now blocked as a search term on the TikTok platform, though using it in the context of a hashtag (#k2spice) during searches is not. Hence, the results of this study are not generalizable to all synthetic cannabinoid discussions or content occurring on the platform.

Although this study identified additional hashtags associated with synthetic cannabinoid use, it did not purposefully examine these hashtags for additional incarceration-related content and did not examine prison-specific hashtags (eg, #PrisonTikTok). Future studies should incorporate new, emerging, and trending substance-related and incarceration-related keywords, code words, and hashtags to better understand the changing social network dynamics of synthetic cannabinoid promotion and behavior. Moreover, as findings are based on a convenience sample generated by the platform’s own search algorithms, the scope of K2/Spice access and abuse among specific marginalized populations warrants additional investigation and inclusion of additional data sources that may be used by incarcerated populations. We were also unable to confirm if the contents described in this study were actually generated in correctional facilities due to the limited metadata available from TikTok accounts and posts.

TikTok is a popular social media platform with millions of daily active users. Our study highlights the growing risk for the public and specific marginalized populations who may interact with and post psychoactive substance–related content on TikTok. Minimizing the potential individual and population-specific harm of this content on vulnerable prison populations should be a priority for platforms and the criminal justice system alike, as well as making attempts to strengthen substance abuse screening policies and evidence-based treatment to better ensure timely rehabilitation and reentry into society.