Joe Dispenza and the Evidence Behind the Retreat

Joe Dispenza fills convention halls. The proposition is unusually specific for the genre: that sustained meditation paired with what he calls elevated emotion can shift gene expression, immune markers, and brain connectivity — and that some people recover from serious illness as a result. Specific claims are testable, which is more than most of this field offers.

The testable part has now been partly tested. A 2025 paper in Communications Biology, from Hemal Patel's group at UC San Diego, followed attendees at a seven-day retreat and reported changes in neural connectivity alongside molecular shifts in blood samples. It is a real study in a Nature Portfolio journal, and it is the strongest published evidence in this space.

What the study can and cannot support

It had no active control group. Participants were self-selected, highly motivated, and paid to attend. A week away from work, in a room of thousands of people expecting transformation, with long hours of sitting practice and altered sleep, changes a great deal about a person's physiology. Nothing in the design separates the effect of Dispenza's specific method from the effect of the retreat itself.

The frequently repeated IgA and heart-brain coherence figures rest on weaker ground still — internal measurements and unpublished EEG work, not independently replicated data. A registered trial of the retreat protocol has been completed but its results were not in the peer-reviewed literature at the time of writing.

A finding produced by an organisation that also sells the intervention is not disqualified. It is simply not yet independent.

The honest position

Meditation has a substantial evidence base for stress physiology, mood, and inflammatory markers. That base does not extend to remission of structural disease, and neither Dispenza's published work nor anyone else's has established it. The interesting question — whether an intense, belief-laden group practice does something a solitary app-based practice does not — remains genuinely open and genuinely under-studied.

What we would want to see next is straightforward: a randomised comparison against an equally intense, equally social retreat using a different method, with pre-registered outcomes and independent labs running the assays. Until that exists, the honest description of the evidence is promising and preliminary, not proven.