Nature-themed video intervention may improve cardiovascular safety of psilocybin-assisted therapy for alcohol use disorder

Heinzerling and colleagues situate their study within renewed interest in psychedelic-assisted therapies for alcohol use disorder (AUD), noting historical and recent positive findings for LSD and psilocybin. Earlier trials suggested single doses of psychedelics can reduce drinking for months, and a recent randomized, placebo-controlled trial found psilocybin plus counselling reduced percent heavy drinking days. The authors emphasise that ‘‘set and setting’’ — the participant's mindset and the physical/social environment — shape psychedelic experiences and outcomes, and that interventions which foster connection with nature have been associated with reduced stress and blood pressure in other healthcare contexts. Despite this, randomised trials testing set-and-setting augmentations for psychedelic therapy are lacking. This pilot study tested whether a purpose-made nature-themed video intervention, ‘‘Visual Healing’’, is feasible, safe, and tolerable when integrated into psilocybin-assisted therapy for AUD, and whether it affects clinical and physiological outcomes. The investigators hypothesised that Visual Healing might reduce pre-session anxiety and post-psilocybin cardiovascular responses, increase feelings of nature connectedness and potentially augment mystical or therapeutic effects, without disrupting the psychedelic experience. The trial therefore compared Visual Healing versus standard set-and-setting (eyes-shades and music) during the first of two open-label 25 mg psilocybin dosing sessions, with choice of procedure for the second session.

This was a pilot randomised, controlled trial enrolling 20 adults with DSM-5 moderate-to-severe AUD. The intended sample size (N = 20) was selected to assess feasibility and fit budgetary constraints for a single-centre pilot. Participants were recruited by community outreach and screened with clinical interviews, medical assessments, urine drug screen, EKG, and alcohol breath tests; exclusion criteria included current major depressive disorder or generalized anxiety disorder, recent serotonergic medication use, significant cardiac disease or uncontrolled hypertension, recent or heavy lifetime psychedelic use, pregnancy, active suicidality, and other standard safety exclusions. One participant discontinued for alcohol relapse during the trial. After screening and baseline assessments, participants were randomised 1:1 to Visual Healing (N = 10) or Standard procedures (N = 10) for the Week 3 dosing session. All participants received two open-label 25 mg oral psilocybin sessions scheduled 4 weeks apart (Week 3 and Week 7), weekly counselling from Weeks 1–10, and a follow-up at Week 14. For the second dosing session participants could choose which procedure to receive. Randomisation was performed via a pre-generated list held by staff without direct participant contact. Visual Healing consisted of three bespoke videos: a 3-min Intro video shown at the end of preparatory sessions, a 42-min Opening Video played at the start of the dosing session (immediately after dosing, during the Ascent phase), and an optional 15-min Closing Video after peak effects had resolved (Descent/Return phase). The Opening and Closing videos used nature cinematography with music, no narration, and were timed to avoid most of the reported peak subjective effects; they were presented on an 85-inch LCD with surround sound while participants reclined. Standard procedures followed typical trial practice: participants reclined, wore eyeshades, and listened to a curated music playlist. All participants received a structured counselling protocol delivered by a two-person clinician team (physician plus therapist), combining preparatory sessions, integration, and AUD-focused counselling modelled on prior trials. Preparatory sessions addressed intention setting, safety, breathing strategies, and optional supportive touch consent. Participants in the Standard group completed a 3-min guided body scan during prep; Visual Healing participants instead viewed the 3-min Intro video. During dosing, clinicians minimised interaction, monitored vitals and adverse events at prespecified timepoints (30, 60, 90, 120 min and 4, 6, 7 h post-dose), and had rescue medications and procedures available. Primary endpoints were feasibility (recruitment/retention and minutes of video viewed), safety and tolerability (adverse events and vital signs). Secondary endpoints included self-reported alcohol use (Timeline Followback; PROMIS Alcohol Use Short Form). Exploratory measures included state anxiety (STAI-SF), salivary cortisol, and psychedelic experience instruments (MEQ30, Ego Dissolution Inventory, Emotional Breakthrough Inventory, Challenging Experience Questionnaire, Questionnaire for Psychotic Experiences). Repeated-measures models were used to compare group, time, and group-by-time interactions, with significance set at p < 0.05 for this pilot.

Of 55 consented/screened individuals, 20 met eligibility and were randomised; 19 of 20 (95%) completed the 14-week protocol. One Standard-group participant was discontinued at Week 6 after a severe alcohol relapse requiring inpatient detoxification; two serious adverse events (ED visit in Week 2 and inpatient admission in Week 6) occurred in that single Standard-group participant and were not temporally related to dosing. Baseline demographics and drinking severity did not differ significantly between groups. Exposure to Visual Healing during the Week 3 session averaged 37.9 minutes of the 42-min Opening Video (4/10 viewed the whole opening video); six participants watched the optional Closing Video averaging 8.3 minutes. When offered in Week 7, uptake was mixed: 6/10 of those initially randomised to Visual Healing chose it again, and 5/9 of prior Standard participants chose to view the video for the first time. There were no statistically significant between-group differences in minutes viewed (p > 0.05). Adverse events across the trial were generally mild to moderate, most frequently headache, elevated blood pressure, and nausea. One participant in the Visual Healing arm vomited at the end of the Opening Video but declined to attribute cause; no other events were clearly video-related and no participants required rescue medication. The Standard group reported a higher frequency of adverse events overall, but this difference did not reach statistical significance (Kruskal–Wallis p = 0.07). Alcohol use declined substantially across the whole sample. Mean drinking days and mean heavy drinking days per week decreased from baseline through the preparatory period and again after the Week 3 psilocybin session, remaining low thereafter. Repeated-measures analyses showed significant time effects for drinking days (e.g. F (6,107) = 7.16, p < 0.0001) and heavy drinking days (F (6,107) = 15.76, p < 0.001). PROMIS Alcohol Use total score decreased over time (F (4,18) = 12.9, p < 0.0001). There were no significant differences between Visual Healing and Standard groups on alcohol outcomes at any time point. Cardiovascular responses to psilocybin rose transiently during dosing but returned to normal by discharge; no serious cardiovascular events occurred. Importantly, the peak pre-to-post psilocybin increase in blood pressure was significantly smaller in participants randomised to Visual Healing versus Standard: systolic BP difference χ2(1) = 4.98, p = 0.02; diastolic BP χ2(1) = 4.03, p = 0.04. Peak heart rate increases were also numerically smaller in the Visual Healing group but did not reach significance (χ2(1) = 2.78, p = 0.10). Measures of the subjective psychedelic experience did not differ between groups during Week 3: MEQ30, Ego Dissolution Inventory, and Emotional Breakthrough Inventory scores were similar (p > 0.05). Overall, 50% of participants met criteria for a complete mystical experience (40% Visual Healing; 60% Standard; Fisher's exact p = 0.7). Challenging Experience Questionnaire totals and subscales tended to be lower in the Visual Healing group, with the Fear subscale approaching significance (χ2(1) = 4.00, p = 0.05). Anxiety decreased over time across both groups (time main effect F (6,18) = 5.28, p = 0.003) but there were no between-group differences (F (6,18) = 1.15, p = 0.4). Morning salivary cortisol showed no significant group differences or time effect (F (2,18) = 1.90, p = 0.18; time F (2,18) = 2.25, p = 0.13). Participant feedback was largely favourable: most agreed Visual Healing increased relaxation/preparedness and feelings of connection to nature. Seventy four percent of responses disagreed or strongly disagreed that Visual Healing distracted from the psychedelic experience; 26% of responses strongly agreed Visual Healing improved the psychedelic experience and 37% strongly agreed they would choose it for future sessions. Representative comments included that videos ‘‘helped ease into the experience’’ and were ‘‘calming & grounding’’ when used at the close of a session, though one respondent described the material as "annoying, cliché."

Heinzerling and colleagues interpret these pilot data to indicate that integrating a nature-themed video intervention into psilocybin-assisted therapy for AUD is feasible, acceptable to participants, and not associated with video-related adverse events. They emphasise that the intervention did not appear to interfere with core psychedelic subjective effects or with clinical reductions in alcohol use, while being associated with smaller peak increases in systolic and diastolic blood pressure during the dosing session. From the authors' perspective, reduced post-psilocybin blood pressure reactivity may reduce cardiovascular risks and thus broaden safe access to psychedelic therapies, particularly in populations where cardiostimulant effects are a concern. The discussion situates these findings relative to earlier work: the magnitude and timing of alcohol reductions were similar to prior psilocybin AUD trials, and MEQ30 scores paralleled published values, despite this trial using a fixed 25 mg dose. The authors note that alcohol use decreased even before dosing, consistent with preparatory counselling and the requirement to abstain prior to dosing. Key limitations acknowledged include the open-label design, lack of blinding to the video intervention, small sample size, and limited diversity (predominantly White sample). The authors caution that the conservative timing and length of the videos were chosen to avoid interfering with peak effects, and they note ethical concerns about delivering suggestive interventions during periods of increased suggestibility following psychedelics. They also recognise they did not collect reasons for declining the Closing Video and that uptake of the Closing Video was variable, limiting inference about its effects. Generalisability to other video content or formats is unknown; the intervention used bespoke nature material produced for the trial. For future work, the investigators call for larger, more diverse, and adequately powered trials to replicate the cardiovascular finding and to explore whether video or virtual-reality interventions at different time points or with tailored nature content can safely augment therapeutic effects. They also highlight the need for careful ethical consideration when delivering interventions during putative windows of increased neuroplasticity or suggestibility.

The authors conclude that embedding the Visual Healing nature-themed video within open-label psilocybin-assisted therapy for AUD was feasible, safe, and well tolerated in this pilot randomised trial. Preliminary evidence suggests Visual Healing may reduce peak post-psilocybin blood pressure increases, increase patient relaxation and preparedness, and improve satisfaction without diminishing mystical or alcohol-related outcomes. They recommend replication and further trials of set-and-setting augmentations, including different video interventions and other psychedelic medications and indications, before broader implementation.