The AI Psych Brief

Effectiveness of rTMS vs. Pharmacotherapy in Late-Life Treatment-Resistant Depression: An Observational Study of Individual Participant Data From Two Randomized Clinical Trials

Adler E, Blumberger DM, Lenze EJ, Brown PJ, Karp JF, Lavretsky H, Rhee TG, Mulsant BH, Kaster TS · American Journal of Psychiatry · October 2026

The authors pooled individual participant data on 758 adults aged 60 or older with treatment-resistant depression from two randomized trials: one testing once-daily bilateral rTMS for 4–6 weeks, and one comparing aripiprazole augmentation, bupropion augmentation, and a switch to bupropion. Because the patients came from different trials, they used restriction, propensity-score weighting, and regression adjustment to make the groups comparable.

At 10 weeks, rTMS was associated with MADRS scores 4.2 points lower than aripiprazole augmentation and about 6 points lower than either bupropion strategy. Response rates favored rTMS as well (adjusted odds ratios of roughly 4 against each medication arm), although the confidence intervals were wide. The main tolerability cost of rTMS was more headache.

Read the paper →  doi:10.1176/appi.ajp.20260398

Dynamic change patterns in suicidal ideation differ in Brief Cognitive Behavioral Therapy (BCBT) and Present-Centered Therapy (PCT)

Bryan CJ, Tabares JV, Khazem LR, Baker JC, Brown LA, Taylor DJ, Pruiksma KE, Acierno R, Rudd MD · Behaviour Research and Therapy · October 2026

BCBT prevents suicide attempts better than present-centered therapy, yet the two produce similar drops in suicidal ideation scores, which is a puzzle. Using session-by-session ratings from a randomized trial in high-risk military personnel and veterans, the authors tracked five separate items of the Scale for Suicidal Ideation (wish to live, wish to die, reasons for living versus dying, suicidal desire, and passive death wish) rather than a single total score.

In both treatments some dimensions moved quickly and others slowly. But in BCBT the dimensions changed faster and more in step with each other, while in PCT change was slower and less coordinated across dimensions. The same total score can hide qualitatively different routes to improvement.

Read the paper →  doi:10.1016/j.brat.2026.105169

SSRI hyponatremia in older adults shows up early, so check sodium early

SSRI- and SNRI-associated hyponatremia is mostly SIADH: increased ADH release or renal sensitivity leaves the patient euvolemic with concentrated urine and a falling serum sodium. The timing is the clinically useful part. Most cases appear in the first two to four weeks of treatment, not months later, so a sodium checked at baseline and again at about two to four weeks catches most of them. Risk stacks: age over 65, low body weight, female sex, concurrent thiazide use, and a baseline sodium already at the low end of normal.

Early symptoms in older adults look like the depression or its treatment going wrong: fatigue, unsteadiness, new confusion, or a fall. Mirtazapine and bupropion carry lower reported risk and are the usual alternatives when an older patient has already had SSRI hyponatremia, though no antidepressant is risk-free. This is one more reason the rTMS result in today's issue matters for late-life depression.

When you are the good therapist and the last one was the bad one, you are being cast

Patients with borderline organization often begin treatment by describing the previous therapist as cold, careless, or harmful, and the new therapist as the first one who really understands. It feels like rapport. In TFP terms it is a split: the patient's internal world is organized around an idealized object and a persecutory one, and you have been assigned the idealized role. The previous therapist is holding the persecutory half. Splits are unstable, so the role you are in now predicts the one you will be in later. When you accept the idealization, you collude with the defense. The devaluation then arrives as a shock and often ends the treatment.

The technical move is to name both halves of the dyad early, while the idealization is still in place, and frame them as the patient's way of organizing relationships rather than a fact about the old therapist. You do not have to defend your colleague. You are describing a pattern that will come for you too, before it does.