Opparounds

Obstetric Risk Associated With Stimulant Versus Nonstimulant Treatment During Pregnancy

Dara E. Babinski, PhD; Wen-Jan Tuan, DHA · Journal of Clinical Psychiatry · August 2026

Using the TriNetX federated electronic health record network, Babinski and Tuan identified nearly 895,000 pregnancies in women 18-45 and split them into three groups: stimulant-exposed (n=3,465), nonstimulant ADHD-medication-exposed (n=575), and unexposed. Compared with no medication, both stimulant and nonstimulant use were associated with higher odds of obstetric complications, with odds ratios roughly 1.2 to 1.9 for stimulants and up to 5.7 for nonstimulants depending on the specific outcome.

Head-to-head, stimulants carried higher odds of placenta previa and large-for-gestational-age infants, while nonstimulants carried higher odds of gestational diabetes, placental abruption, intrauterine growth restriction, and preterm delivery. This is a retrospective EHR cohort, so confounding by ADHD severity and indication is a real limitation the authors themselves flag.

Read the paper →  doi:10.4088/JCP.26m16372

Evaluation of Interventions for Cognitive Symptoms in Long COVID: A Randomized Clinical Trial

David S. Knopman, Deborah Koltai, and the RECOVER-NEURO Consortium · JAMA Neurology · November 2025

RECOVER-NEURO randomized 328 adults with cognitive Long COVID lasting 12+ weeks to one of five 10-week, fully remote arms: online puzzles (active control), adaptive computerized cognitive training (BrainHQ), structured cognitive-behavioral rehabilitation (PASC-CoRE) plus BrainHQ, or active or sham transcranial direct current stimulation, each paired with BrainHQ.

On the primary outcome — the modified Everyday Cognition Scale — none of the active interventions outperformed the puzzle-and-games control; adjusted between-group differences were 0.0 to 0.1 points, and every arm improved by a similar 0.4 to 0.5 points over the 10 weeks. Secondary neuropsychological testing told the same story. The strongest predictor of improvement turned out to be time since infection, not treatment assignment.

Read the paper →  doi:10.1001/jamaneurol.2025.4415

Lithium doesn't just affect the seizure — it affects the paralytic

Lithium potentiates the neuromuscular blockade produced by succinylcholine, the depolarizing paralytic used for the induced seizure in ECT — the mechanism is thought to involve lithium's effect on presynaptic acetylcholine release and membrane sodium conductance, on top of lithium's own weak anticholinesterase-like activity. The practical result is a prolonged block and delayed return of spontaneous respiration after an otherwise routine ECT session, not a seizure-threshold problem.

Most protocols manage this by holding the morning lithium dose, using a reduced succinylcholine dose (or switching to a non-depolarizing agent) on treatment days, and running train-of-four monitoring longer than usual before extubation. The trap is assuming a stable outpatient lithium level from last week still applies — NPO status and dehydration around the ECT visit can push the level up on the day it matters most, exactly when the interaction is least forgiving.

Negative capability is a discipline, not a delay

Bion borrowed Keats's phrase negative capability — being able to remain in uncertainty, mystery, and doubt without an irritable reaching after fact and reason — to describe a specific analytic stance: the capacity to sit with material you don't yet understand instead of resolving the discomfort by naming it.

The pull to interpret early is usually the therapist's anxiety about not knowing, not the patient's readiness to hear something. An interpretation offered to relieve your own uncertainty forecloses a meaning that hasn't finished forming; the patient learns the room isn't safe for things that are still inchoate. This is not passivity — it's active tracking, attention, and restraint, held until the material is ready to be named, ideally by the patient rather than for them.