Psychiatry
Semaglutide for alcohol use disorder missed its primary endpoint but cut heavy drinking days
A small phase 2 trial is a reason to keep watching GLP-1 drugs, not to prescribe them for alcohol. A sham-controlled circadian app reduced mood episode recurrence, and peers' diagnoses tracked adolescents' own later diagnoses.
Therapy that plans the return to work was linked to fuller return to work
Work-focused CBT was associated with fuller return to work after sickness absence; remote therapy for women showed moderate effects on depression; and cognitive gaps in early-onset psychosis looked larger with age.
Social support and problem-solving carry task-shared therapy; relaxation may drag it down
An individual-participant network meta-analysis dismantles lay-delivered psychological care into its parts; excess COVID-19 deaths in severe mental illness outlasted the vaccine roll-out; and glutamate marks antipsychotic non-response.
Physical illness accumulates two to three times faster in psychosis and bipolar disorder — and earliest in the young
A 178,665-person cohort makes the case for physical health checks from the first episode, the Cochrane update finds no clear advantage for zuclopenthixol acetate over standard rapid tranquillisation, and long-term psychodynamic therapy earns a strong recommendation for depression.
Escitalopram and amitriptyline lengthen QTc most, but trials show no early cardiac events from antidepressants
An individual-participant analysis of double-blind trials ranks ten antidepressants by QTc effect, an observational review links antidepressants to ventricular arrhythmia, cannabis is tied to violence both committed and suffered, and two trials — Open Dialogue and paediatric anxiety sequencing — report null primary outcomes with useful secondaries.
Semaglutide and metformin lead the evidence for antipsychotic weight gain
A network meta-analysis ranks weight-loss drugs in schizophrenia; ketamine outperformed midazolam in treatment-resistant bipolar depression; rapid relapse after stopping paliperidone looks like illness, not withdrawal; and GLP-1 agonists were associated with lower mortality in serious mental illness.
The symptoms that do not move, and the mood stabiliser that does
Collaborative care shifts most of depression but not suicidality; partners resemble each other more than anyone assumed; and lithium and valproate come out level on suicide risk.
Long-acting injectables cut manic relapse in bipolar disorder
Hong Kong's within-person data put psychiatric admission down by a third on depot antipsychotics, with the benefit concentrated in mania and mixed states. Plus what is left two years after NMDA receptor encephalitis, and the dentate gyrus finding that belongs to ECT alone.
Semaglutide misses its primary outcome in alcohol use disorder and still looks interesting
A phase 2 trial in treatment-seeking drinkers, the case against dopamine as a single final pathway in psychosis, what falling polygenic scores say about rising ADHD and autism rates, and an app that cut mood recurrence threefold.
Two in five trafficking survivors have PTSD, and a fifth have attempted suicide
A global meta-analysis quantifies a burden most services never screen for. Alongside it: what 'neurodivergent' actually denotes in the literature, a gut-signal marker that predicts anorexia relapse, peer effects on adolescent risk, and a sober look at low-dose lithium in mild cognitive impairment.
Anxious distress, and the subgroup that gained most from augmentation
A post hoc look at adjunctive lumateperone found the effect nearly twice as large in patients with anxious distress. Plus what a 12-year psychosis follow-up says about the first three years, trauma-focused therapy in low-income settings, and why half the mental health apps you recommend will be gone in two years.
Vagus nerve stimulation misses its primary endpoint and still looks useful
RECOVER's 493-patient sham-controlled trial in markedly resistant depression failed on its prespecified outcome but favoured active stimulation on symptoms, function and quality of life; virtual reality beat a short CBT course for persistent voices; and excess COVID mortality in severe mental illness never came down.
In task-shared therapy for depression and anxiety, social support and problem-solving did the work — relaxation did not
An individual-participant network meta-analysis of lay-delivered therapy closes the edition. Before it: physical multimorbidity accumulating early in psychosis and bipolar disorder, Open Dialogue in crisis care, home tDCS, and sequencing treatment for anxious children.
Semaglutide and metformin lead for antipsychotic weight gain, and ketamine helps treatment-resistant bipolar depression
A 95-trial network meta-analysis ranks the options for antipsychotic-associated weight gain; a midazolam-controlled trial supports ketamine in bipolar depression without switching to mania; and paliperidone discontinuation data question the idea of rebound relapse.
Suicide prevention rethought from India, and exercise earns its place in perinatal care
A JAMA Psychiatry viewpoint argues that global suicide policy is built on evidence from countries where most suicides do not happen. Plus rTMS for exposure-resistant OCD, aromatherapy for menopausal mood, and a meta-analysis that puts a number on exercise in pregnancy.
Depot antipsychotics earn their place in bipolar disorder
A twenty-year Hong Kong record linkage halves manic relapse on long-acting injectables; prescribable depression apps deliver a real but narrow effect; and a Danish registry shows how much of what we call heritability comes from who people partner with.
Executive function is not what is holding back AUD treatment
In moderate alcohol use disorder without much comorbidity, cognitive testing found no impairment at all and predicted nothing about who cut down. Plus esketamine at induction and postoperative mood, a workplace app with a dropout problem, and test anxiety in half of all medical students.
Who your patient sits next to, and what their HbA1c is doing
Six hundred thousand Finnish adolescents show peer environments tracking mental health a decade on; esketamine fails a postoperative anxiety trial; and a 26-study meta-analysis puts a mortality figure on depression with diabetes.
Semaglutide missed its primary endpoint in alcohol use disorder and cut heavy drinking days anyway
A 50-patient phase 2 trial found no effect on cue-elicited craving but fewer heavy drinking days, fewer drinks per drinking day and less cannabis use; a Danish cohort shows the polygenic burden of newly diagnosed autism and ADHD has fallen year on year; and adjunctive lumateperone improved rather than worsened sexual function in resistant depression.
Two-thirds of NMDA receptor encephalitis patients carry symptoms into the post-acute years
Seventy patients a median of 22 months from onset show persistent memory and affective problems mapped onto reduced hippocampal and fronto-cingulo-temporal complexity; suicide-focused psychotherapy improves pain and social functioning as well as ideation; and half of mental health apps disappear within five years.
Every infection severe enough for admission leaves a psychiatric shadow
A million matched pairs show raised two-year psychiatric and neurological risk after hospitalisation with infection at any body site, worst after encephalitis. Also: excess COVID mortality in severe mental illness never came down, and where to aim TMS for smoking cessation.
Relaxation training made people worse, and a circadian app cut mood episode recurrence threefold
A component network meta-analysis of 10,612 participants dismantles task-shared psychosocial therapy and finds social support does the work while relaxation does harm, and a sham-controlled trial of a circadian feedback app prevents relapse in mood disorders.
Ketamine worked in treatment-resistant bipolar depression without tipping anyone into mania
A midazolam-controlled trial found a 7.3-point MADRS advantage over two weeks with no mania, hypomania or psychosis in either arm, and a network meta-analysis of 95 trials ranks semaglutide and metformin as the most effective drugs for antipsychotic-induced weight gain.
Ketamine works in treatment-resistant bipolar depression without triggering mania
Ket-BD randomised 68 outpatients with treatment-resistant bipolar depression to four ketamine or midazolam infusions over two weeks. MADRS fell 7.3 points further with ketamine, and no case of mania, hypomania or psychosis occurred in either arm. Plus semaglutide leads the field for antipsychotic-induced weight gain.
Physical illness stacks up decades early in severe mental illness
A 178,665-person Queensland cohort shows people with schizophrenia or bipolar disorder accumulate multi-system physical disease far younger, a large database study links GLP-1 agonists to lower mortality in this group, and rapid relapse after stopping paliperidone is not a withdrawal effect.