Releasing anxiety, and what the trials show

Anxiety is the one corner of this literature where the evidence is not thin. Music therapy has been studied for decades across hospitals, surgical wards and emergency departments, and the recent reviews have started to converge on the same shape.

The broadest read is a systematic review with multilevel meta-analyses by de Witte, Aalbers, Vink and colleagues, published in eClinicalMedicine in 2025. Pooling across clinical populations, the authors describe music therapy as an effective and scalable intervention for anxiety symptoms. One detail matters for a room like ours: receptive and combined approaches, meaning listening rather than performing, outperformed active only methods.

Scale is part of why that finding matters. Anxiety care is expensive and slow to reach people, and a review that finds a listening based intervention working across clinical populations is describing something that can be delivered to a room rather than one person at a time. It does not replace care. It sits alongside it.

Two trials, two populations

Angkoontassaneeyarat and colleagues, writing in the International Journal of Emergency Medicine in 2025, ran a randomized trial at Ramathibodi Hospital. Patients who received music therapy alongside standard analgesia reported significantly lower anxiety and pain scores than patients who received analgesia alone. An emergency department is a hard place to feel calm, and the effect still showed up there.

Ranjan, Jaiswal, Mallik and colleagues, in Cureus the same year, followed forty four postsurgical cancer patients through two weeks of adjuvant music therapy. Anxiety scores came out lower than in standard care, a mean difference of 1.73 at a p value of 0.040. That is a modest, real effect in a group of people with every reason to be anxious, and it is the kind of number worth quoting exactly rather than rounding up into a headline.

It is worth knowing what these scores are. Anxiety in these trials is measured with questionnaires, usually taken before and after, and the differences are modest by design. Nobody in this literature is claiming an hour of music empties a person out. The effects are small, repeated across enough studies to be credible, and they stack with everything else a person is already doing.

Duration did something intensity did not

Mullen, Peng, Stewart, Mallik and Russo published a randomized trial in PLOS Mental Health in 2026 in which music combined with auditory beat stimulation produced greater reductions in anxiety and negative affect than a control condition, following a dose response pattern. The thirty six minute condition produced the largest drop in negative affect, significantly larger than the twelve minute one.

We take two things from that body of work and leave the rest alone. Listening, done properly, is not passive. And time on the clock matters more than most people expect, which is why a session here runs sixty minutes and not fifteen.

Listening turns out to be the active ingredient.

What putting it down feels like

Anxiety lives in a loop. A thought lifts the body, the lifted body confirms the thought, and the loop pulls tighter. Sitting still with it often feeds it, because a still body with nothing else to attend to attends to itself.

The room interrupts that from the outside. Low frequency sound gives the nervous system something large and physical to follow that is not you. It is genuinely difficult to keep rehearsing a conversation from Tuesday while a bass note is moving through your ribs. Attention goes to the biggest signal in the space, and for an hour the biggest signal is warm, slow and outside your own head.

Then there is the company. Seven other people are lying in the dark doing exactly the same thing, close enough that you can hear them breathe, far enough that nobody is going to speak to you. For a lot of our guests that combination, together and unbothered, is the part that lands hardest.

None of that makes a session treatment, and we would not describe it as one. What it makes is an hour in which the loop has nothing to run on.

Feel Water is the quiet entry session, sixty minutes, from $63. Come alone. Most people do the first time.