Psychiatry

Cumulative prenatal adversities predict persistent youth psychopathology and cortical thinning

Among 8515 ABCD youth, 78% had at least one adverse prenatal exposure (APE). Cumulative APE burden showed a strong dose–response association with clinically significant CBCL total problems across four years (OR ~2 for 1 APE; ~7 for ≥3). Associations with ADHD symptoms weakened over time, whereas links with depressive symptoms strengthened, suggesting evolving risk profiles. Greater APE burden was associated with accelerated cortical thinning in many regions and persisted in sibling-contrast analyses. These findings support counseling about clustered prenatal risks and early monitoring for internalizing symptoms in highly exposed children. 1

LF‑rTMS to left temporoparietal junction best supported for TRS auditory hallucinations

This network meta-analysis synthesized 55 RCTs of 33 non-invasive brain stimulation modalities in 1981 patients with treatment-resistant schizophrenia. Low-frequency rTMS over the left temporoparietal junction yielded robust improvements in auditory hallucinations versus sham (SMD −0.784, 95% CI −1.202 to −0.366). Neuronavigated continuous theta burst stimulation of left temporoparietal cortex also improved positive symptoms and hallucinations. Several neuronavigated theta burst protocols targeting left dorsolateral prefrontal cortex improved negative and overall symptoms. All modalities appeared acceptable, but heterogeneity and low confidence for some outcomes argue for cautious, individualized implementation. 2

AI-generated feedback improves engagement, not mean depression scores, in self-guided iCBT

This three-arm RCT randomized 1187 adults with depressive symptoms to AI-augmented self-guided iCBT, identical iCBT without AI, or waitlist. AI support did not significantly change mean PHQ-9 scores versus standard iCBT at seven weeks or three months. However, AI-iCBT participants had higher odds of weekly participation over time (OR 1.23, 95% CI 1.09–1.39). Exploratory analyses linked activation of empathic AI feedback with stronger engagement and antidepressant effects. Clinically, AI augmentation currently appears more useful for adherence than for enhancing average symptom reduction. 3

Behavioral activation yields moderate–large effects for adolescent depression, especially in school settings

This meta-analysis pooled 11 RCTs (572 adolescents) testing behavioral activation (BA) for subthreshold and major depression. BA showed a moderate effect versus treatment-as-usual (SMD −0.42) and a large effect versus no-treatment controls (SMD −0.87). Effects were larger for mild than severe depression and were particularly strong in university or school settings. Interventions without parental involvement outperformed those with parents, likely confounded by age, severity, and setting. BA also moderately improved activation and functioning, supporting brief, school-based BA as a pragmatic first-line option for older youth. 4