Between April 2021 and February 2026, 220 trained peer supporters answered 25,419 contacts to a US psychedelic support line. That record is now the basis of what researchers say is the largest dataset ever assembled on real-world psychedelic support interactions.
Published in Science Advances, the Fireside Database offers researchers a view of experiences that often happen far from clinics. Psychedelics are used in homes, at festivals, in ceremonies and in other community settings, but evidence about what people encounter there has been limited. The database gathers information about substances, distress, social settings, reasons for seeking help and the kinds of support provided. It also includes demographic details when callers chose to share them, plus 2,777 post-interaction satisfaction surveys with open-ended answers.
Psilocybin was the most commonly reported substance, appearing in 59.9% of interactions coded by substance. Cannabis appeared in 12.8% and had the highest overall rate of distress. But the researchers found that cannabis users were more likely to call while still intoxicated. When they compared similar types of calls, distress did not differ significantly among the major substances. Whether a person sought help during an experience or afterward was a stronger clue to distress than the substance reported.
That finding could help shape crisis-support and harm-reduction training across different drug categories. The data may also let researchers study whether differences between peer supporters affect callers’ experiences, and whether those lessons can improve training for therapists, facilitators and other support workers.
The database also records 43 interactions classified as reports of facilitator abuse. Such reports were linked disproportionately with ceremonial settings and plant medicines. The authors say this is a signal of possible harm in settings that are not regulated and deserves more study. Fireside did not independently investigate or verify the individual reports.
The authors stress that the records cannot show how common distress is among all people who use psychedelics or cannabis. They describe callers who chose to contact a support line, not a representative sample of drug users. The distress rates in the database therefore should not be treated as estimates of overall risk.
Fireside Project founder and executive director Joshua White, a co-author, said the resource was built over more than five years because researchers have had little way to study the settings where most psychedelic use takes place. Lead author Felix Schoeller said clinical trials offer important evidence under controlled conditions, but cover only a small part of real-world use.
The de-identified interaction and survey records, along with the code used to reproduce the paper’s figures and statistics, are being made available for research. The authors hope the material will support work in public health, policy, safety monitoring and practitioner training. By opening a clearer view of support calls—and their limits—the database gives future studies a stronger starting point for asking what helps people stay safer when they seek help.