The Potential Economic and Public Health Impact of MDMA-Assisted Group Therapy for PTSD in Ukraine

Marseille and colleagues frame the study within the substantial mental health burden created by Russia's military aggression against Ukraine, noting large numbers of civilians exposed to trauma and high estimated prevalence of PTSD and complex PTSD. Earlier research and health-system experience suggest existing pharmacological and psychotherapeutic options are frequently inadequate: roughly half of PTSD patients fail to achieve meaningful improvement with standard care. The authors cite Phase 3 trial results showing marked efficacy for MDMA-assisted therapy (MAT) in individual formats and argue that adapting MAT to a group format may be necessary in Ukraine because of severe shortages of trained mental health professionals. This study updates a previously published decision-analytic Markov model to evaluate the costs, health outcomes, and broader societal effects of a modified group-based MAT protocol with supplemental individual sessions in a Ukrainian setting. The investigators model outcomes for a notional cohort of 1,000 PTSD patients, report results from healthcare payer and limited societal perspectives over 1-, 3-, and 5-year horizons, and project scaled-up impacts if 25%, 50% or 75% of eligible civilians receive MAT over a 10-year roll-out. Key outcomes include per-patient MAT costs, quality-adjusted life years (QALYs) gained, premature deaths averted, net costs or savings, and net monetary benefit (NMB).

The analysis uses an annual-cycle Markov model that categorises patients into five PTSD severity states (asymptomatic, mild, moderate, severe, extreme) and follows transitions through these health states until death. The model’s severity distribution at baseline mirrors that observed in 90 participants from a Phase 3 randomised trial; treatment effectiveness is represented as the observed shift in severity distribution from pre- to post-treatment in that trial measured about eight weeks after the final MDMA session. Mortality rates by severity were taken from published sources and scaled to Ukraine using age-specific background mortality as a referent. Health state utilities were derived from EQ-5D-5L data collected in the Phase 3 trial and converted to utility weights for QALY calculation. Costs and QALYs are discounted at 3% per year. Costing used a micro-costing approach based on Ukrainian compensation rates, local costs for screening/laboratory tests, and an estimate for MDMA acquisition costs; a 10% overhead addition was applied to variable costs. Excess healthcare costs associated with PTSD were estimated by regressing per-patient PTSD costs in a 2020 European systematic review on country per-capita health spending and applying the result to Ukraine’s adjusted per-capita spending for 2022. Productivity and caregiver costs were included in the two limited societal perspectives, and a third societal perspective additionally incorporated economic losses from premature mortality. The base-case assumes MAT efficacy persists for 3 years; results are also reported for 1- and 5-year efficacy durations. The modified treatment protocol costed for Ukraine departs from the MAPS Phase 3 individual format and combines group and individual elements to expand capacity. Each patient receives six 60-minute preparatory sessions (three individual, three group), three group-based MDMA dosing sessions (groups of six patients with two therapists), and 12 integration sessions (three group, one individual after each MDMA session), totalling 21 therapy sessions. The model excludes military personnel and assumes scale-up scenarios in which 25%, 50% or 75% of eligible civilians are treated, with 10% of the target population served each year over a 10-year period; incident PTSD cases and implementation-period mortality were not modelled. Sensitivity analyses (univariate and multivariate stochastic) were conducted using @RISK software, and a Ukraine-specific willingness-to-pay (WTP) threshold was estimated at $5,501 per QALY by scaling thresholds from other countries using GDP ratios.

Base-case findings for a cohort of 1,000 patients, assuming 3 years of MAT benefit, indicate net discounted healthcare payer costs of $1.1 million, 717 QALYs gained, and 19.2 premature deaths averted. Under a WTP of $5,501 per QALY, the healthcare payer analysis produced a net monetary benefit (NMB) of $2,843 and an incremental cost-effectiveness ratio (ICER) of $1,537 per QALY in the 3-year scenario. Extending the assumed duration of benefit to 5 years improved cost-effectiveness (NMB $5,716; ICER $595), whereas assuming benefit lasted only 1 year yielded an NMB near zero and an ICER of $5,261 per QALY, which just meets the estimated WTP threshold. When productivity gains and reductions in caregiver costs were included (limited societal perspective), the analysis showed MAT could generate net savings. The authors report societal savings of $2.6 million per 1,000 patients assuming 3 years of benefit; in those broader perspectives MAT in many scenarios was the dominant option (both less costly and more effective). For scaled-up implementation, the model projects that treating 50% of eligible civilians over 10 years could save about 48,000 lives, gain roughly 1.5 million QALYs, and deliver net societal savings on the order of $5.6 billion (reported as a central estimate). Sensitivity analyses identified the cost of MAT and estimates of productivity and caregiver effects for severe PTSD as the most influential inputs. For the 1,000-patient analysis (perspectives including productivity), multivariate stochastic analysis produced a 90% confidence interval for NMB of roughly $3.6–$6.1 million (payer plus productivity) and $5.2–$8.0 million (including premature mortality losses). For the 10-year, 50% uptake scaled scenario, multivariate analyses yielded wide ranges: estimated net costs varied from -$4.3 billion (net savings) to $0.3 billion (90% CI); estimated net savings ranged approximately $3.0–$8.4 billion in multivariate runs. Estimated undiscounted lives saved over 10 years varied from 15,000 to 88,000 (90% CI). Finally, probabilistic results indicated MAT exceeds a 50% probability of being cost-effective at WTP values above about $1,500 per QALY, and approaches near-certain cost-effectiveness by $2,500 per QALY for the base-case perspective; the two broader societal perspectives reach near-certain cost-effectiveness at much lower WTP levels.

Marseille and colleagues interpret their modelling as indicating that introducing a group-adapted MAT protocol in Ukraine would likely be highly cost-effective from a healthcare payer perspective and cost-saving from a limited societal perspective, while producing substantial public health gains. They emphasise that a group-based MAT format reduces per-patient clinician time and thereby increases treatment capacity—an important consideration given Ukraine’s shortage of mental health professionals. The authors situate their findings alongside earlier Phase 3 trial evidence showing strong clinical efficacy for MDMA-assisted therapy in individual formats and cost-effectiveness studies from higher-income settings, arguing that a modified group format can preserve many economic and health advantages while improving scalability. The investigators acknowledge several important limitations. Key model inputs—PTSD mortality risks, severity distribution, and MAT efficacy—largely derive from U.S. data, including a single Phase 3 trial that used individual rather than group therapy; direct evidence for group MAT efficacy is limited. Ukrainian patient populations may differ in trauma type, demographics, and cultural factors in ways that could affect real-world effectiveness. The analysis excludes military personnel, omits incident PTSD cases during the 10-year implementation window, and does not capture many wider societal effects (for example, reduced domestic violence, accidents, or criminal justice involvement). The authors note that productivity benefits from reduced disability were not fully captured, implying their estimates may be conservative with respect to total societal benefit. Finally, they flag implementation challenges—training providers and integrating MAT into existing services—that would need to be addressed if policymakers were to pursue adoption. On policy and research implications, the authors contend the economic and public-health case for MAT in Ukraine is compelling and robust across plausible scenarios, but call for implementation-oriented research to assess real-world effectiveness, feasibility of group delivery, workforce training needs, and context-specific outcomes in the Ukrainian setting.

The study concludes that making MDMA-assisted therapy available for PTSD in Ukraine would probably be cost-effective from the healthcare payer perspective and cost-saving when broader societal impacts are considered, while producing large gains in QALYs and averting substantial numbers of premature deaths. Marseille and colleagues recommend that policymakers consider MAT as part of a comprehensive mental health response, while recognising the need for further research on implementation and outcomes specific to Ukraine and for addressing workforce and service-integration challenges.