Depression Trials and TMS Marketing: Comparing Like With Like

Two TMS services can advertise very different success rates without necessarily treating comparable patients or measuring the same outcome. One may quote remission immediately after a course, another response several weeks later, and a third an internal estimate among completers. The numbers become useful only after those differences are visible. Comparing like with like is a research skill that can improve a consultation without requiring statistical expertise.

Identify the claim before checking the number

Ask whether the service is claiming that TMS exceeds sham, that one protocol is comparable to another, or that a specific clinic achieves a particular outcome. Those are different questions. A randomized sham-controlled trial addresses a causal comparison under defined conditions. An active-comparator trial addresses relative performance. A clinic registry describes its practice population but may have no equivalent control group.

The George et al., sham-controlled prefrontal rTMS trial, 2010 reported remission in 14.1% of the active group and 5.1% of the sham group in its intention-to-treat analysis during the masked phase. Those figures belong to a specific historical protocol and selected population. They should not be used as a universal modern clinic benchmark, but neither should the sham result disappear when the study is cited to support efficacy.

Match the population

Look for diagnosis, age range, prior treatments, current medicines, and major exclusions. Evidence in nonpsychotic unipolar depression does not automatically estimate outcomes in bipolar depression, psychotic depression, or another disorder. The fact that a device can deliver stimulation does not establish efficacy for every condition. If a service treats a broader population than the cited study, ask how that difference affects uncertainty and clinical oversight.

Treatment resistance is also not a uniform background characteristic. Studies may define it differently, and people may differ in episode duration, severity, and previous care. A percentage from a less resistant sample should not be attached without qualification to a more complex patient group. Matching does not require identical biographies, but it does require acknowledging the differences most likely to affect interpretation.

Match the intervention

The Blumberger et al., THREE-D randomized noninferiority trial, 2018 evaluated iTBS against 10-Hz rTMS using a defined noninferiority design. The Cole et al., Stanford Neuromodulation Therapy randomized trial, 2022 tested a different package combining individualized targeting and high-dose accelerated iTBS against sham. Both involve TMS, but they do not validate every schedule using the same general terminology. The target, pulse structure, spacing, number of sessions, and outcome timing all help identify the actual intervention.

Ask which study most closely matches the clinic's protocol and which components differ. A service should not use a highly favorable result from a specialized protocol while delivering a materially different one without explanation. Our protocol evidence guide offers a structured way to make that comparison. Technical detail should clarify the claim rather than overwhelm the patient into accepting it without scrutiny.

Match the endpoint and follow-up

Response usually refers to a specified symptom reduction; remission means falling below a specified threshold. Neither automatically establishes recovery of functioning, and an end-of-course result does not establish durability. Ask for the exact scale, criterion, assessment date, and any additional treatment received before follow-up. Terms such as success, recovery, or life-changing improvement need definitions before they can be compared with trial outcomes.

The SNT trial's primary outcome was measured four weeks after treatment, not simply at the end of the last day. That timing is part of the result. If a clinic quotes a different follow-up interval, it should explain why. A graph of average symptom reduction can also hide substantial individual variation, so ask about nonresponse and discontinuation rather than focusing only on the most favorable summary.

Examine the clinic's denominator

For internal results, ask whether the percentage includes all people who began, only completers, or only those who returned a survey. Find out what happened to people with missing outcomes. A testimonial page is not a representative denominator. People who did not benefit, stopped because of burden, or were lost to follow-up may be less visible than enthusiastic completers, and that can distort an apparent success rate.

Use the TMS course guide to put outcome estimates alongside safety, cost, and reassessment arrangements. For compressed schedules, the accelerated checklist adds practical comparisons. A responsible service can explain why it believes treatment is reasonable without promising a personal result. The goal is not to choose the largest percentage; it is to choose a decision process grounded in relevant evidence and honest limits.

Medical disclaimer: This article provides general education, not diagnosis, individualized treatment advice, or emergency care. Evidence and regulatory statements refer to the dated sources cited, not an exhaustive verification of all changes through October 30, 2026.

Sources

1. George et al., sham-controlled prefrontal rTMS trial, 2010
2. Blumberger et al., THREE-D randomized noninferiority trial, 2018
3. Cole et al., Stanford Neuromodulation Therapy randomized trial, 2022