Researchers have turned to the video website YouTube to better understand the phenomenon of “robo-tripping” — ingesting large amounts of Robitussin or other cough medicine to become intoxicated.
“This paper demonstrates information about commonly ingested doses of DXM and their side effects can be retrieved from YouTube comments using techniques from information retrieval and natural language processing,” Michael Chary of the Ichan School of Medicine at Mount Sinai and his colleagues wrote in their study, which was published by PLoS One.
“Analyzing social media could provide data to answer questions about the recreational use of substances that would be difficult to obtain through other means.”
Dextromethorphan is a cough suppressant found in numerous medications, including Robitussin and Coricidin Cough and Cold. At high doses, the drug produces dissociative hallucinatory effects similar to ketamine and PCP. The recreational use of DXM has sharply increased over the past few years, according to the researchers
By using computational linguistics to analyze comments on YouTube, the researchers found the most commonly ingested amounts of DXM ranged from 0 to 1500mg. Death was associated with dosages above 1800mg.
The comments also showed that distinct symptoms occurred at  four specific ranges of dosages. These different recreational dose ranges are referred to by users as “plateaus.” The first plateau produces mild stimulant effects, the second produces a drunk- or stoned-liked experience, the third produces dissociative out-of-body experiences, and the fourth produces a full-blown dissociative condition, the researchers wrote.
Data from the YouTube comments strongly resembled documented case reports of DXM intoxication, suggesting accurate information about recreational drug use can be gleaned from the popular video website.
“This study investigated whether YouTube is a useful source of information on the recreational use of an over the counter substance whose usage is, otherwise, challenging to track,” Chary and his colleagues wrote. “If data from social media about recreational drug use concord with clinically documented symptoms and doses, then those data could be used to explore aspects of recreational drug use, such as short-term trends, that are mostly inaccessible with current means: case reports from emergency rooms and poison center calls.”
