How do response and remission rates of repetitive transcranial magnetic stimulation compare between late-life and midlife depression in randomized trials?

How well does rTMS actually work in older adults?

The strongest evidence comes from pooling many randomized trials that compared real rTMS against a sham version, where the coil is placed so the magnetic pulse does not meaningfully stimulate the brain. Across ten such trials in older adults, active rTMS produced a moderate improvement on the Hamilton Depression Rating Scale, a standard clinician-rated symptom score, and the odds of responding were about 2.6 times those of sham, while the odds of remitting were about 6.9 times higher [1]. A separate pooling of nine trials found similar direction and size: response odds about 2.9 times sham and remission odds about 4 times sham [2].

Those odds ratios sound enormous, and it helps to read them as comparisons, not as personal probabilities. An odds ratio of 4 for remission means the odds of remission were four times higher on active treatment than on sham; it does not mean 400 percent or even 80 percent of patients remitted. The trials also varied in dose, coil placement, and patient selection, which is why the two pooled estimates differ somewhat in size [1][2].

One practical detail stands out: the benefit was concentrated in trials using stimulation intensity of at least 100 percent of the resting motor threshold, the minimum magnetic dose needed to make a thumb twitch. In that group, symptom improvement and response odds were clearly better than sham, while trials below that intensity showed no significant difference from sham [1]. That pattern suggests dose matters, though it comes from subgroup comparisons rather than a dedicated head-to-head dose trial.

Why is there no clean late-life versus midlife comparison?

The randomized evidence base is built around older adults specifically. The pooled trials enrolled patients with late-life depression, generally meaning later adulthood, and compared active against sham stimulation within that group [1][2]. None of these trials set out to randomize midlife and late-life patients side by side and compare their response and remission rates.

That means any claim that older patients respond worse, or just as well, than midlife patients cannot be settled by these trials. The comparison the question asks for would require trials that enroll both age groups under the same protocol and report outcomes separately, and that design is not represented here. What the randomized data can support is a statement about how well rTMS performs in late life relative to placebo, not relative to younger patients.

Within older adults, who is most likely to remit?

Even inside late-life samples, remission is uneven, and the strongest signal is about timing in the treatment pathway. In one randomized trial of 164 older adults comparing standard bilateral rTMS with a faster theta-burst form, remission was about 44 percent in people who had tried at most one prior medication, compared with roughly 26 to 27 percent in those who had already failed two or more [3]. A larger analysis of the same trial found that lower baseline depression severity, fewer prior antidepressant failures, and higher global cognition each predicted remission [5].

The cognitive finding is worth pausing on. Higher scores on a brief cognitive screen, the Montreal Cognitive Assessment, were associated with greater likelihood of remission, and the statistical model flagged a combination of milder symptoms, at most one prior medication failure, and a cognition score of 23 or higher as a particularly favorable profile [5]. This is a prediction from observed data, not proof that cognition causes remission, and it should not be read as a reason to withhold treatment from patients with lower scores.

Together these results point the same way: rTMS appears to work best when used earlier, before multiple medication failures accumulate [3][5]. That is a statement about who tends to benefit most within late-life samples, not evidence that late-life patients do worse than midlife patients.

Is it tolerable, and what would settle the age question?

Efficacy only matters if people stay in treatment. In the pooled comparison of active versus sham rTMS in older adults, discontinuation for any reason and dropout due to adverse events did not differ significantly between the two groups, which supports the view that rTMS is well tolerated in this population [2].

Several questions remain genuinely open. The pooled analyses themselves call for larger samples and longer-term outcome data before the evidence can be considered robust [1]. A registered trial of a priming approach, in which high-frequency stimulation precedes low-frequency stimulation over the right prefrontal cortex, is designed to test whether that combination helps in late-life depression, but as a protocol it reports no response or remission results yet [4]. Most importantly, the late-life versus midlife comparison remains an evidence gap: the randomized trials establish that rTMS outperforms sham in older adults, and they identify earlier treatment and milder, less cognitively impaired presentations as favorable, but they do not measure how those rates stack up against midlife patients treated the same way [1][2][5].

Sources used in this answer

Related research questions