This Week in Psychiatry — Oct 5, 2026

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Psychiatry. This week we're covering 10 notable papers spanning psychotropic safety and who benefits from our medications, the rapidly expanding field of brain stimulation and its biomarkers, and new evidence on augmenting psychotherapy and on the trajectory of childhood mental illness. Let's dive in.

We start with pharmacotherapy, where several large datasets this week look at who benefits from our drugs and at what cost. In The Lancet Psychiatry, Furukawa and colleagues pooled individual participant data from 18 fixed-dose trials, covering about seven thousand adults with acute schizophrenia treated with haloperidol, olanzapine, paliperidone, risperidone, or placebo [1]. The headline is reassuring and fairly firm: there was no evidence that antipsychotics work differently in women and men on overall symptoms, and the dose-response curves were essentially identical, plateauing at around 3 to 5 milligrams of risperidone equivalents in both sexes. Two differences did emerge. Antipsychotics reduced all-cause dropout more in women than in men, a finding the authors rate as moderately credible, and prolactin elevation was substantially greater in women, rated as highly credible. Women made up less than a third of participants, and only five drugs were studied, so the findings cannot be stretched to newer agents. Still, the authors conclude that existing efficacy estimates apply to both sexes, while the prolactin signal adds weight to monitoring for hyperprolactinaemia in female patients.

In PLOS Medicine, McIlvride and colleagues linked primary care, maternity, and education records for nearly 170 thousand children born in Wales to examine prenatal exposure to maternal depression and antidepressants [2]. Both exposures were associated with a higher likelihood of special educational need by school age. In absolute terms, about one in five unexposed children had a special educational need, rising to roughly one in four among children exposed to both depression and antidepressants. Antidepressant exposure was also linked to higher odds of autism and to a near doubling in the rate of later ADHD diagnosis. No single antidepressant class stood out clearly. The authors are explicit about the central limitation, confounding by indication: women prescribed antidepressants likely had more severe depression, and depression itself carried excess risk. This is observational evidence that refines the absolute risk discussion in perinatal counselling, but it cannot separate the drug from the illness, and it does not by itself argue against treating depression in pregnancy.

A second PLOS Medicine paper, from Chen and colleagues, reaches into oncology [3]. Preclinical work has suggested the serotonin transporter acts as an immune checkpoint on T cells. Using a target trial emulation in a large electronic health record network, the team compared roughly fifteen hundred matched pairs of patients with solid tumours and depression or anxiety who were taking either an SSRI or a benzodiazepine when they started immune checkpoint inhibitor therapy. Two-year mortality was about 23 percent with SSRIs versus about 34 percent with benzodiazepines, roughly a third lower hazard of death, and every individually analysable SSRI showed a similar association. Immune-related adverse events were modestly more frequent with SSRIs, driven by thyroid dysfunction, and negative control outcomes were null. The authors flag the key weakness plainly: performance status was not recorded, and benzodiazepines are often prescribed in very different circumstances, so residual confounding is a real possibility. This is hypothesis-generating work that the authors say warrants prospective trials, not evidence that changes psychotropic choice in oncology today.

Rounding out this theme, Cruz and colleagues in the American Journal of Psychiatry analysed more than a billion benzodiazepine fills in the United States between 2013 and 2024 [4]. Annual prevalence of benzodiazepine dispensing fell by about a quarter, to roughly five and a half percent of the population, with the steepest relative decline, nearly half, among young adults aged 18 to 29. Dispensing fell sharply among pain specialists, primary care physicians, and psychiatrists, and mean dose per prescription declined across the board. The exception was advanced practice providers, whose benzodiazepine dispensing roughly doubled. The authors read the overall decline as greater caution, likely linked to safety warnings and labelling changes, and identify the shift toward advanced practice providers as an area for ongoing monitoring.

Our second theme is neuromodulation, which Sackeim and George describe in the American Journal of Psychiatry as a Cambrian explosion [5]. Their review surveys how multiple forms of brain stimulation have gained Food and Drug Administration clearance across neuropsychiatric indications, and points to emerging directions: noninvasive focal drug delivery that bypasses systemic side effects, deep targeting without affecting surrounding tissue, closed-loop stimulation guided by personalised biomarkers, and home-use devices that have now received clearance. It is a narrative overview rather than new data, but it frames the week's original studies well.

Also in the American Journal of Psychiatry, Song and colleagues retrospectively analysed 131 patients with treatment-resistant depression who received subcallosal cingulate deep brain stimulation across three independent cohorts [6]. Responders had a more consistent spatial pattern of stimulated tissue than nonresponders, and stimulation mapping pointed to a focal cluster linked to improvement. Critically, failure to activate the left cingulum bundle and the medial uncinate fasciculus weakened the therapeutic effect. The authors argue this supports connectome-based targeting over anatomical landmarks alone. The design is retrospective and relies partly on normative connectome data, so prospective validation is still needed, but it helps explain the inconsistent results in earlier deep brain stimulation trials for depression.

On the biomarker side, Xie and colleagues in Biological Psychiatry meta-analysed 21 studies combining transcranial magnetic stimulation with EEG, covering about a thousand patients with major depression [7]. Negative-polarity responses, especially the N100 component, were modestly enhanced in depression and tracked symptom severity. Treatment dampened the N100, and that reduction was strongly coupled with remission. Baseline signals also appeared to predict response, though the authors label that evidence preliminary. These are pooled small studies, and the measure is far from routine clinical use, but it is a coherent signal across diagnosis, monitoring, and prognosis.

Finally in this theme, a double-blind sham-controlled trial in Molecular Psychiatry from Tian and colleagues tested 15 daily sessions of theta-frequency transcranial alternating current stimulation in 56 men with methamphetamine use disorder [8]. Active stimulation improved proactive cognitive control and inhibition on a stop-signal task, enhanced the relevant frontal midline theta activity, and restored coupling between that brain rhythm and task performance. The outcomes were laboratory measures of cognition rather than drug use or relapse, the sample was small and male only, so this is mechanistic proof of concept rather than a treatment ready for addiction services.

Our last theme turns to psychotherapy augmentation and to children. In Psychotherapy and Psychosomatics, Muehlbacher and colleagues randomised 97 inpatients with obsessive-compulsive disorder to high-intensity or low-intensity exercise immediately after twice-weekly exposure and response prevention sessions over eight weeks [9]. The high-intensity group improved by about three more points on the Yale-Brown scale at discharge, a moderate-to-large effect that persisted at six months. Responder rates were about 85 percent versus about 61 percent, although remission rates did not differ, and depressive symptoms also improved more at follow-up. This is a single inpatient trial, so generalisation to outpatient care remains untested, but it supports exercise as a feasible, low-cost adjunct to exposure therapy and extends similar findings from anxiety and trauma disorders.

And in the Journal of Child Psychology and Psychiatry, Armitage and colleagues used three national probability surveys in England, from 1999, 2004, and 2017, each with three-year follow-up, to ask whether outcomes for children with a psychiatric diagnosis have changed [10]. They have worsened. Having a disorder at baseline predicted more difficulties and greater functional impact at follow-up in every wave, but the association grew stronger with each survey. Among children with a disorder in 1999, about a third deteriorated over three years, compared with about half of children with a disorder in 2017. The authors cannot yet say why, but the finding suggests rising prevalence is accompanied by poorer trajectories, not just more diagnosis.

If you only have time for one paper this week, make it the individual participant data meta-analysis on sex differences in antipsychotic response from The Lancet Psychiatry [1]. It largely settles a long-running question about whether women and men need different antipsychotic dosing, while sharpening attention on prolactin as a sex-specific harm.

Here is what this week's evidence adds up to in Psychiatry. First, antipsychotic efficacy and dose-response appear equivalent across sexes for the older and second-generation agents studied, with prolactin elevation the main sex-specific difference, a conclusion resting on strong pooled trial data. Second, prenatal antidepressant exposure is associated with modestly higher absolute rates of special educational need, autism, and ADHD, but confounding by depression severity remains unresolved, so the causal question is still open. Third, the survival association between SSRIs and immunotherapy outcomes is intriguing yet observational, and only prospective trials can confirm it. Fourth, brain stimulation is maturing toward precision targeting and biomarker guidance, with connectome-based deep brain stimulation and TMS-EEG signatures showing promise that still awaits prospective validation. And fifth, children with psychiatric disorders in England appear to be faring worse over time than earlier generations, while exercise after exposure therapy offers one early, trial-based route to strengthening psychological treatment in adults with obsessive-compulsive disorder.

That's your roundup for This Week in Psychiatry. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.