The cost-effectiveness of MDMA-assisted psychotherapy for the treatment of chronic, treatment-resistant PTSD

This study (2020) on the costs (and benefits) of MDMA-assisted therapy for PTSD finds it to be more cost-effective than other treatments. It’s based on the data from six double-blind, placebo-controlled phase II trials (n=105) done by MAPS.


Background Chronic posttraumatic stress disorder (PTSD) is a disabling condition that generates considerable morbidity, mortality, and both medical and indirect social costs. Treatment options are limited. A novel therapy using 3,4-methylenedioxymethamphetamine (MDMA) has shown efficacy in six phase 2 trials. Its cost-effectiveness is unknown.

Methods and findings To assess the cost-effectiveness of MDMA-assisted psychotherapy (MAP) from the health care payer’s perspective, we constructed a decision-analytic Markov model to portray the costs and health benefits of treating patients with chronic, severe, or extreme, treatment-resistant PTSD with MAP. In six double-blind phase 2 trials, MAP consisted of a mean of 2.5 90-minute trauma-focused psychotherapy sessions before two 8-hour sessions with MDMA (mean dose of 125 mg), followed by a mean of 3.5 integration sessions for each active session. The control group received an inactive placebo or 25–40 mg. of MDMA, and otherwise followed the same regimen. Our model calculates net medical costs, mortality, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios. Efficacy was based on the pooled results of six randomized controlled phase 2 trials with 105 subjects; and a four-year follow-up of 19 subjects. Other inputs were based on published literature and on assumptions when data were unavailable. We modeled results over a 30-year analytic horizon and conducted extensive sensitivity analyses. Our model calculates expected medical costs, mortality, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratio. Future costs and QALYs were discounted at 3% per year. For 1,000 individuals, MAP generates discounted net savings of $103.2 million over 30 years while accruing 5,553 discounted QALYs, compared to continued standard of care. MAP breaks even on cost at 3.1 years while delivering 918 QALYs. Making the conservative assumption that benefits cease after one year, MAP would accrue net costs of $7.6 million while generating 288 QALYS, or $26,427 per QALY gained.

Conclusion MAP provided to patients with severe or extreme, chronic PTSD appears to be cost-saving while delivering substantial clinical benefit. Third-party payers are likely to save money within three years by covering this form of therapy.”

Authors: Elliot Marseille, James G. Kahn, Berra Yazar-Klosinski, Rick Doblin


This paper is included in our ‘Top 10 Articles Introducing Psychedelic-Assisted Therapy

The data on the participant outcomes, comes from Mithoefer et al (2019) which analysed the outcomes of the MAPS phase II trials.

This paper is also discussed in a press release by MAPS.

In the US, PTSD affects 6.8% of people during their lifetime and 3.5% of the population in the last year (Kessler et al, 2005a; Kessler et al, 2005b). Women, on average, are twice as likely to suffer from PTSD than men. Bothe and colleagues (2020) estimates the costs related to PTSD (medical costs) at €43.000 whilst Von der Warth and colleagues (2020) estimate the costs at between $5.500 and $24.450 for both direct and indirect costs. The current study estimated the medical cost at $19.899 for severe PTSD.

The present study investigates if MDMA-assisted psychotherapy may be a possible cost-effective intervention to alleviate these costs and the suffering experienced by those with PTSD.

The participants, on which the cost-estimates are based, are 105 people suffering from PTSD, of which 74 were in the active treatment group. They suffered from moderate to extreme chronic PTSD for an average of 16.5 years. The mean CAPS-IV score at baseline was 84 on average.

The study used a Markov model to simulate the costs and benefits. In this study, this meant they used it to model cost and benefits into the future with (pseudo)randomly changing data such as remission rates (as so to come to an average over X numbers of simulations).

The costs of treatment were estimated at $7,543. Of this $6.194 (82%) were the costs for therapists, $997 (13%) for screening, and $353 (5%) for the lab-grade MDMA.

  • The costs for therapists could potentially fall quite significantly if a less (over)qualified (second) therapist could hold/support the sessions, and/or if group therapy (see Anderson et al, 2020) was investigated/implemented.

The benefits of the study were estimated at 25% of the medical costs, for the proportion for which the intervention was successful (versus a Markov model of the same patients but at their baselines, i.e. without the intervention).

The results from the 10.000 simulations indicate an average (projected over 30 years) saving of 43 deaths averted, 5.500 added qualitative years (QALYs), and $103 million saved, per 1000 people treated. Or 5.5 added QALYs per person, and $103.000 medical care costs saved per person. The break-even (for costs) is estimated to be at 3 years after the intervention.

The paper then also tackles different variations/estimates for the costs and resulting benefits.

The model used in this study had several limitations. On the benefit side, it did not include benefits to the family and friends (added QALYs, lower domestic abuse, substance abuse, and productivity gains) and broader society. On the costs side, the study assumed no further relapses, a reasonable assumption for this treatment, but possibly less applicable to depression (TRD, MDD) for which a variety of psychedelics (e.g. psilocybin) in combination with therapy are being investigated.

Outside what the authors mention, there can be limitations in the applicability of these results and the effectiveness of this type of therapy for PTSD. Others (Goth & Hoeijmakers, 2020, p38) have pointed out that 1) the therapy standards could be higher in the study versus when scaled up, 2) study volunteers are more likely to respond than the average PTSD case (especially since the severity was so high in this study group), 3) being part of a study can lead to social desirability bias or observer effects (Hawthorne effect), and 4) risks of these results not replicating outside of these studies (although the number of treated patients is high in this study, it’s still only 74).

When looking at the cost for each added qualitative year (QALY), the current paper estimated this at 5.5 QALYs per $6.194 or $1.126 per QALY. This is significantly lower than the $28.000 to $42.000 (converted from pounds) the UK is willing to spend on medical interventions per QALY (NICE, 2013).



PTSD is a serious psychiatric condition that may follow a traumatic event. It can cause stress-mediated physical health problems, including cardiovascular disease and type-2 diabetes, as well as alcohol abuse, high caloric intake and BMI, and smoking.

Interviews with 9,282 American adults indicated a lifetime prevalence of PTSD of 6.8% and past-year prevalence of 3.5%. Women were more than twice as likely as men to suffer from PTSD.

Against this backdrop, recent evidence of benefit from MDMA-assisted psychotherapy (MAP) is particularly relevant. The non-profit Multidisciplinary Association of Psychedelic Studies (MAPS) is working with the Food and Drug Administration and the European Medicines Agency to build on pooled efficacy data from six phase 2 trials.


We developed a decision analytic model to portray clinical benefits, medication costs, and net medical costs for 1,000 patients with PTSD. The model was based on six phase 2 clinical trials conducted between 2004 and 2017.

Patient population

The 105 subjects of six double-blind controlled pilot studies suffered moderate to extreme chronic PTSD with a mean age of 40.5 years, and had failed at least one conventional therapy for PTSD.

Treatment protocol

Following recruitment, randomization, and two to three non-drug 90-minute therapy sessions, participants received placebo/control or active doses of MDMA administered during two 8-hour psychotherapy sessions conducted 3 – 5 weeks apart.

Representation of clinical trial results

The Markov model captures the intervention effect as the difference in PTSD severity following the intervention versus baseline. The control group showed no improvement or remission, which is consistent with the stable clinical status of patients who, by trial inclusion criteria, suffer from chronic PTSD.

In the phase 2 trials, patients received psychotherapy combined with either a placebo (two trials) or a 25 – 40 mg dose of MDMA believed to be clinically inactive. The active treatment group experienced some improvement, so in a sensitivity analysis an additional comparison was implemented.

Health state utility values

We assigned participants to mild, moderate, or severe PTSD according to estimates from the Global Burden of Disease, and used the Quality of Well Being-Self-Administered Scale to define extreme PTSD.

Medical care costs

We estimated the medical costs of patients with PTSD from four studies with five separate estimates, and assumed that severe PTSD would cost $19,899 annually. Asymptomatic cases cost $4,949 annually.

PTSD is associated with higher mental health and general medical care costs. The model conservatively distributes the reduction in health care costs over five years.


PTSD is associated with elevated mortality. The relative risk for PTSD patients is 2.76, 2.51, 2.05, and 1.74, respectively, using age-specific background U.S., mortality rate as the referent.

Analytic time horizon

Our analysis projects MAP benefits beyond 4 years, assuming retention of clinical benefits, and finds the cost breakeven point at 3.1 years.

Sensitivity and scenario analysis

We conducted extensive sensitivity analyses to assess variation in findings given parameter value uncertainty. We projected MAP costs and benefits assuming that patients progressed to the next most severe stage of PTSD at a rate of 6% annually.

Base-case results

The cost of the MAP intervention was $7,543 per patient. This includes therapists’ compensation, MDMA, test kits, and carotid ultrasound.

MAP averts 42.9 undiscounted deaths, generates 5,553 discounted QALYs, and saves a discounted net $103 million compared to controls. Using a 10-year horizon, MAP saves 2,517 QALYs, averts 18.9 deaths, and saves $36.7 million.

Sensitivity and scenario analyses

In one-way sensitivity analyses, MAP has an ICER of $2,124 per QALY gained at 10% of base-case PTSD medical costs. Variation in effectiveness has the greatest down-side effect on the magnitude of savings.

In a two-way sensitivity analysis, MAP breaks even at a cost of $16,232 per patient and has an ICER of $2,779 per QALY gained.


MDMA-assisted psychotherapy appears to be cost-saving and thus highly cost-effective, generating 5,553 QALYs and averting 42.9 deaths for every 1,000 patients treated.

Many third-party payers are likely to save money by including MAP as a covered benefit for patients with chronic PTSD. The risk of losing members is modest because the break-even time is 3.1 years and patients must migrate to another plan at an average annual rate of 33%.

This study, which draws on a substantial body of RCT data on clinical effectiveness, suggests that a treatment for PTSD could reduce disability payments and raise productivity. However, the study’s model excludes several important potential benefits to families and society.

MAPS is currently planning a 5-15 year follow-up study of patients from the phase trials 2 to shed further light on durability of MAP benefits.

The data on medical care costs associated with PTSD are highly variable, and the trials were only conducted on a limited portion of PTSD patients in the U.S.

We are aware of no other studies of the cost-effectiveness of MDMA psychotherapy. It is possible to reduce costs by conducting some of the psychotherapy sessions in a group context, and by using a master’s level practitioner with specialized additional training.

MDMA-assisted psychotherapy can alleviate PTSD symptoms in a large portion of patients with the most severe and treatment-resistant forms of the disorder.


Authors associated with this publication with profiles on Blossom

Elliot Marseille
Elliot Marseille is a health economist specializing in psychedelics and HIV/AIDS, focusing on cost-effectiveness in global health challenges.


Institutes associated with this publication

Global Initiative for Psychedelic Science Economics
The Global Initiative for Psychedelic Science Economics (GIPSE) is a network of health economists dedicated to achieving the potential of psychedelic therapies for high-priority mental health conditions. 

Linked Research Papers

Notable research papers that build on or are influenced by this paper

Updated cost-effectiveness of MDMA-assisted therapy for the treatment of posttraumatic stress disorder in the United States: Findings from a phase 3 trial
This study (2022) builds on previous research assessing the cost-effectiveness of MDMA-assisted therapy (MDMA-AT) for the treatment of PTSD by assessing the data from a recent phase III trial. MDMA-AT as conducted in the phase III trial costs $11,537 per patient. Compared to the standard of care for 1,000 patients, MDMA-AT generates discounted net health care savings of $132.9 million over 30 years. Ultimately, MDMA-AT for severe or chronic PTSD is cost-saving while delivering substantial clinical benefit.

PDF of The cost-effectiveness of MDMA-assisted psychotherapy for the treatment of chronic, treatment-resistant PTSD