CDP highlighted an MDMA therapy review and Army suicide-risk prediction research. Here is how to interpret the findings and their implications for care.
For immediate danger, call 911. For crisis support, call 988 and press 1 for the Veterans Crisis Line. You can also text 838255.
On October 22, 2026, the Center for Deployment Psychology (CDP) published a Research Update highlighting an MDMA-assisted therapy review and an Army study of suicide-related risk prediction. CDP is part of the Uniformed Services University of the Health Sciences. The roundup listed research on PTSD treatment outcomes alongside research using data available at periodic health assessments (PHAs).
The update brings together two distinct areas of military mental-health research. It is a research roundup, not a clinical guideline or a new treatment standard.
The studies address different questions that matter to military and veteran healthcare. The MDMA review examines treatment outcomes for people with post-traumatic stress disorder (PTSD). The Army study examines whether administrative information available during health assessments can help identify soldiers at comparatively higher risk of later suicide-related outcomes.
Treatment research asks whether an approach changes symptoms or functioning. Prediction research asks whether available information helps distinguish groups with different levels of future risk. Those questions belong in the same healthcare conversation, but evidence answering one does not answer the other.
For active personnel, the Army findings relate directly to research involving Regular Army health assessments. For veterans and former service members, the broader relevance is how emerging evidence might inform future treatment discussions and risk identification. The studies did not test a combined program linking MDMA-assisted therapy with prediction models.
Within the broader picture of military and veteran health, the practical value is understanding exactly what each study measured. A treatment result and a prediction score require different interpretations before either becomes useful in care.
The review listed by CDP examined MDMA-assisted therapy in relation to PTSD symptoms, dissociation, depression and functional disability. That range matters because symptom severity and everyday functioning are separate outcomes. The review's title identifies the questions under study rather than supplying an individual treatment recommendation.
Related evidence comes from a different 2026 meta-analysis published in Neuropsychopharmacology. That analysis reported a reduction in PTSD symptoms compared with control conditions across six studies and 242 participants. Its reported statistical results were:
The same meta-analysis reported higher response and remission rates. These findings belong to that separate analysis and must not be attributed to the review featured in CDP's roundup.
The statistics summarize a comparison between trial groups. The effect size is not a percentage improvement and should not be read as one. Keeping that distinction clear helps prevent a technical result from becoming an exaggerated claim about expected recovery.
The treatment evidence also has a defined boundary: results from trials do not establish effectiveness for every patient or setting. Neither the featured review nor the separate meta-analysis determines whether MDMA-assisted therapy is appropriate for a particular service member or veteran. That personal question requires assessment by a qualified clinician.
The Army study available through PubMed Central analyzed 1,932,243 PHAs completed by Regular Army soldiers from 2015 through 2019. Researchers used administrative data available at the time of each assessment to develop machine-learning models. Outcomes were assessed over the following 24 months.
The reported count refers to assessments, not a count of unique soldiers. The study evaluated what could be learned from information already available during those assessments rather than testing a new treatment.
In the test sample, the authors reported:
Discrimination describes how well a model separates outcomes within the study data. These values are not an individual soldier's probability of a future event. The distinction is important because a model's performance across a dataset is different from a prediction about one person.
Secondary reporting by U.S. Medicine described the assessments associated with higher predicted risk:
The report said the flagged PHAs accounted for roughly 31% of subsequent suicides and 40% of attempts. It also described only partial overlap between assessments flagged for the different outcomes. Suicide risk and suicide-attempt risk therefore should not be treated as interchangeable model outputs.
Those percentages describe how later outcomes were distributed across study-defined groups. They do not describe the share of outcomes prevented. Identifying a group for possible follow-up and demonstrating that follow-up improves outcomes are separate research tasks.
The authors noted relatively low outcome counts compared with the dataset's size. The study counted attempts documented in medical records and could therefore miss undocumented events. Its diagnostic codes also prevented researchers from distinguishing self-directed violence with and without intent to die.
The historical Regular Army findings do not establish performance in every service population or a later operational setting. The study also does not demonstrate that an intervention prompted by a model's risk flag reduces suicide or attempts. Its contribution is evidence about risk identification, not proof of prevention or confirmation that the model is in routine use.
This update supports more precise conversations about care rather than a self-directed change in treatment. For PTSD care, the useful next step is discussion with a qualified clinician about the current evidence and established options. The review can inform that discussion without becoming a personal treatment plan.
Questions supported by the research include:
These questions keep the focus on clinical relevance. A finding becomes more useful when the person receiving care and the clinician understand which problem it addresses. A headline about PTSD symptoms does not answer every question about recovery.
For readers also weighing sleep and mental-health concerns, the same outcome-specific approach is useful. Treatment discussions can distinguish the problems a study measured from other concerns that still need attention. This avoids treating a broad label such as recovery as though it were a single research outcome.
For suicide-related concerns, access to timely support takes priority over interpreting a research score. A current safety concern should not be deferred until a future health assessment or model result. The crisis contacts at the beginning of this article provide immediate support options.
In a clinical setting, a prediction score is best understood as a possible aid to assessment rather than a diagnosis. The relevant care question is what further assessment and support a person needs. Personal safety decisions belong with qualified professionals who can assess the individual rather than relying on a dataset alone.
No exercise prescription or physical-performance testing change follows from this roundup. Its practical implications concern mental-health evidence and clinical conversations. Keeping that scope clear prevents an emerging research finding from being stretched into unrelated health advice.
A useful priority for future treatment research would be clearer evidence about how symptom changes relate to everyday functioning. For prediction research, an important next question is what happens after a person is identified for follow-up. Better separation of risk groups is valuable to study, but the care provided afterward needs its own evaluation.
These remain separate research tracks. Evidence about an MDMA-assisted therapy outcome should be judged on its treatment findings. Evidence about an Army prediction model should be judged on its identification performance and the consequences of using it in care.
For military and veteran healthcare, the most meaningful progress would connect research measurements with practical outcomes that matter to patients. That means asking not only whether a study produced a measurable signal but whether the evidence supports better decisions and timely support.
As these research tracks develop, which findings will translate into better functioning and more timely mental-health care for service members and veterans?
Follow BattleVet for practical guidance on military and veteran health, strength, recovery, testosterone, sleep and healthy aging. Stay connected for new articles, research backed insights and clear information to help you stay capable for the years ahead.
Hims expanded its testosterone program. Here are the product-status, monitoring and fertility questions relevant to veterans comparing care.
An Air Force veteran’s Apple Watch alert prompted medical evaluation. Learn what the reported case means for sleep testing and clinical care.
The Fort McClellan veterans health study is planning its analysis. Learn who is covered, how the research will work, and what it means for care.
Build better habits around strength, recovery, sleep, hormones and healthy aging with practical guidance for active military personnel and veterans.
Explore BattleVet