Group psychedelic therapy: empirical estimates of cost-savings and improved access

Clinical trials and renewed public interest have driven a resurgence in research on psychedelic-assisted therapies for disorders that are often difficult to treat. Marseille and colleagues note that regulators have given Breakthrough Therapy designations to MDMA for PTSD and to psilocybin for major depressive disorder, and that expanded access programmes and local policy changes have further increased momentum. At the same time, the authors identify practical barriers to broad and equitable access—most notably the potential high costs of treatment, uncertain payer willingness, workforce shortages of clinicians trained to deliver psychedelic therapies, and barriers for underserved populations. This study aims to quantify one proposed service-delivery response: incorporating group sessions into psychedelic therapy protocols. Using empirical data from two contemporary trial sites that already include group-based elements, the investigators estimate clinician time, clinician compensation, and overall variable costs per patient for group versus individual session formats. They then project how adopting group modalities could affect clinician workforce requirements and the time or cost needed to treat the population of U.S. adults estimated to be eligible for MDMA or psilocybin therapy.

The analysis used micro-costing data obtained from two trial sites with different drugs, indications and group approaches. One source was the Social Neuroscience & Psychotherapy (SNaP) Lab's MDMA-assisted group therapy protocol for veterans with PTSD; the other was Sunstone Therapies' psilocybin protocol for people with cancer and comorbid major depressive disorder. The two protocols differ substantially: SNaP Lab includes four 90-minute preparatory sessions, two 8-hour open-label MDMA sessions and eight 90-minute integration sessions, with a cohort size of six patients for most group sessions and a clinician team combining doctoral psychologists, a nurse practitioner or peer counselor, and psychiatrist supervision. Sunstone's protocol comprises two 90-minute preparatory sessions, one 8-hour psilocybin administration (25 mg) and two 90-minute integration sessions; its “group” element is supervisory monitoring of three to four concurrently treated patients via closed-circuit television while each patient remains in a separate room with a Master's-level therapist. Personnel costs were estimated by multiplying clinician hourly wages plus fringe by the hours required per patient under each protocol; direct service hours were increased by 20% to capture variable indirect activities such as billing and charting. Intake, screening and laboratory costs were assigned values based on Medicare CPT reimbursements and Quest lab charges. Drug dosing assumptions reflected protocol practice: SNaP Lab patients were assumed to receive two MDMA sessions with 120 mg plus a 60 mg supplemental dose each time (360 mg total); Sunstone patients received a single 25 mg psilocybin dose. Because final medication prices are unknown, the authors used base-case and sensitivity price points: for MDMA $9,000 per treated patient (base) and $1,800 (sensitivity), corresponding to per-milligram scenarios they describe; for psilocybin $1,500 (base) and $500 (sensitivity). To estimate population impact, the team derived numbers of U.S. adults eligible for MDMA and psilocybin therapies using prevalence adjustments reported elsewhere. For PTSD, they applied PTSD prevalence to the adult population, retained the 50% estimated to be severe and chronic, and subtracted 22% for likely disqualifying comorbidities. For MDD, they began with 14.8 million adults with a past-year episode, applied 71.0% to capture those receiving treatment, selected 30% to represent those who failed or relapsed, and then removed 26% for disqualifying comorbidities. Workforce projections were produced by calculating clinician-hours per patient for each protocol, converting those to patients treatable per full-time equivalent (FTE) clinician per year, and then estimating the number of FTEs required to treat all prevalent eligible cases if 10% of the population were treated annually over 10 years. An alternative scenario considered treating with a fixed supply of 5,000 FTE clinicians. Costs over the 10-year horizon were calculated by multiplying treated patients by clinician cost per patient and discounting at 3% annually. The authors report conducting sensitivity analyses varying group size, number of simultaneously monitored patients, and medication price assumptions. They limited the analysis to variable costs and did not include fixed overhead, capital refurbishment, additional training, incident cases, or mortality.

Both group formats reduced therapist-hours and clinician compensation per patient relative to fully individualised sessions. For SNaP Lab's MDMA/PTSD protocol, base-case estimates indicated a saving of 36.2 clinician-hours and $3,467 per patient; this corresponded to a 50.9% reduction in clinician costs and a 20.7% reduction in overall variable costs. For Sunstone's psilocybin/MDD protocol, group monitoring saved 9.4 clinician-hours and $981 per patient, a 34.7% reduction in clinician costs and a 19.0% reduction in overall variable costs (base-case values). Medication and clinician compensation were the largest cost components under the base-case price assumptions. In SNaP Lab's group protocol, MDMA and clinician compensation made up 68% and 25% of total variable costs, respectively, whereas in the equivalent individual-patient configuration MDMA and clinician shares were 54% and 41%. For Sunstone's group model, psilocybin and clinician compensation constituted 36% and 44% of total variable costs; the individual-patient equivalent was 29% and 55%, respectively. Sensitivity analyses showed that the relative savings from group delivery depended on medication prices and group size. If MDMA cost $1,800 per treated patient rather than the base $9,000, the reduction in variable costs from the group format would rise to 36.4% (assuming six patients per group). If psilocybin cost $500 rather than $1,500 per treated patient, savings would rise to 23.5% (assuming four patients monitored). Increasing group size produced modest additional savings: expanding SNaP Lab group size from six to eight patients increased the reduction in variable costs from 20.7% to 22.4%, an extra $277 saved per patient. Operationally, group approaches would allow an additional 28.9 (SNaP Lab) and 29.5 (Sunstone) patients to be treated per FTE clinician annually. When applied to the estimated U.S. prevalent populations, the group model would reduce the FTE clinicians required to treat all eligible patients once over 10 years by 6,711 for the SNaP Lab MDMA/PTSD protocol and by 1,159 for the Sunstone psilocybin/MDD protocol. The extracted text does not clearly report all results for the scenario using 5,000 FTE clinicians, nor does it present detailed numerical totals for the multi-year discounted cost projections in the main text provided here.

Marseille and colleagues interpret their findings as evidence that incorporating group sessions into psychedelic-assisted therapy protocols can substantially reduce clinician time and compensation per patient and thereby expand the effective supply of trained clinicians. They argue this could be important for timely and equitable access because demand for psychedelic therapies after regulatory approval may exceed clinician supply, and current behavioural health wait times and workforce shortages are already substantial. The authors emphasise that reductions in relative costs from group formats depend heavily on the eventual prices of MDMA and psilocybin; when medication costs are large relative to personnel, the proportional effect of group-delivered care on total variable costs is smaller even though clinician costs fall markedly. The paper acknowledges several important limitations. The analysis focused on cost differences and did not evaluate comparative clinical effectiveness or safety between group and individual protocols, and no randomised trials directly comparing these formats exist in the modern literature. Projections of clinicians needed to treat prevalent cases were approximate because they excluded incident cases and deaths, and the costing exercise considered only variable costs—omitting fixed overheads, refurbishment, and potential additional training requirements. The authors also recognise uncertainty about how many patients would accept group modalities; some patients may prefer individual privacy while others may benefit from shared experiences. They point to the Sunstone model (separate rooms with shared remote supervision) as a hybrid that preserves privacy while yielding personnel savings. The investigators note that optimal clinician-to-patient ratios and acceptable group sizes likely vary by psychiatric indication and by the pharmacology of the compound—MDMA may permit larger groups than psilocybin because its effects are described as more predictable. Finally, they observe that if multi-session MDMA protocols (for example, three-session regimens used in Phase III trials) become standard, the cost-containment and workforce-extension benefits of group sessions would become even more consequential. In public-health terms, the authors conclude that group modalities, where clinically feasible, could halve clinician costs for the SNaP Lab MDMA/PTSD protocol (50.9%) and reduce clinician costs by 34.7% for the Sunstone psilocybin/MDD protocol, with more modest but meaningful reductions in overall variable costs (20.7% and 19.0% respectively under base-case drug-price assumptions). They caution that group therapy requires specialised practitioner skills and that quality and safety should not be compromised in the pursuit of cost containment. No new clinical effectiveness claims beyond the reported cost and capacity estimates are made.