Opparounds

Heart Rate Variability Biofeedback for Substance Use Disorder: A Randomized Clinical Trial

Eddie D, Nguyen M, Zeng K, et al. · JAMA Psychiatry · December 2025

This phase 2 trial randomized 120 outpatients with substance use disorder to an eight-week course of heart rate variability biofeedback (HRVB) added to treatment-as-usual, or treatment-as-usual alone, then tracked craving and use in real time via ecological momentary assessment. Among the 115 patients analyzed, HRVB reduced negative affect and craving and was associated with a 64% reduction in alcohol or drug use days (odds ratio 0.36).

The effect appeared to work in part by disrupting the usual link between a craving spike and the use that follows it, rather than simply reducing how often cravings occurred.

Read the paper →  doi:10.1001/jamapsychiatry.2025.2700

Longitudinal Risk for Suicidal Self-Directed Violence Among Veterans With Cancer

Sullivan DR, Disher N, Rosa WE, et al. · JAMA Oncology · July 2026

This national VA cohort followed 292,271 veterans with cancer from 2014 to 2023 and identified 2,400 suicidal self-directed violence events, a rate of 203 per 100,000 person-years. Risk was highest with severe frailty, advanced-stage cancer, comorbid chronic mental illness, and high pain scores.

Critically, risk did not cluster only around diagnosis — it persisted for years into survivorship instead, well past the point most services stop screening.

Read the paper →  doi:10.1001/jamaoncol.2026.1459

The Two-Hour Esketamine Window Isn't Arbitrary — It's a Curve

The REMS-mandated two-hour post-dose monitoring for intranasal esketamine isn't a defensive-medicine buffer; it maps onto the actual time course of the two things that can go wrong. Dissociation peaks within the first 30 to 40 minutes and then reliably fades, so a patient who is intact at 40 minutes rarely deteriorates later in the window. Blood pressure follows a similar arc — transient elevations after dosing, with the REMS protocol flagging intervention only for systolic ≥180 or diastolic ≥110 — meaning most of what you see is expected physiology, not an emergency.

The clinical judgment call is knowing that "still dissociated at 90 minutes" is a genuinely different signal than "dissociated at 35 minutes," even though both look identical on a checklist, because one is on-curve and the other is not. Discharge criteria should track that trajectory — stable and trending down — rather than a single vitals snapshot at the two-hour mark.

When Mentalizing Breaks Down, Rewind the Tape — Don't Push Forward

MBT's stop-rewind-explore technique addresses a specific moment: the patient has stopped mentalizing — dropped into psychic equivalence or pretend mode — and the temptation is to keep talking through it, hoping the conversation finds its way back. That temptation is the error.

The technique is to interrupt the content entirely, name that something just happened ("stop"), and go back to the precise moment before the break occurred ("rewind") rather than the general topic. Only then does the therapist explore what was felt right at that point — the specific thought, image, or interaction that triggered the collapse. This works because mentalizing capacity is state-dependent and moment-specific; asking "why do you feel this way generally" after the break has already occurred asks a mentalizing question of a patient who has, in that instant, stopped being able to mentalize.