Spravato · Relapse detection
Catch Spravato relapse between REMS sessions.
Spravato is dosed under supervision, but the REMS program is silent on the days in between, which is exactly where relapse starts. Continuous monitoring turns that unmonitored window into an early-warning system for esketamine patients.
The gap the REMS does not cover
REMS defines in-clinic monitoring on dosing days: supervised administration, a minimum two-hour observation, and blood-pressure checks. It says nothing about the days between sessions. Across induction and into maintenance, those between-session days are where early relapse signals appear, and they are invisible to the clinic in standard care.
50%
of Spravato patients relapse by 6 months without durable monitoring
79.6%
discontinue within 12 months
ESKALE, France
48h
early-warning lead time from passive signal
How Emobot detects Spravato relapse early
Cover the between-session days
Continuous passive monitoring fills the gap the REMS leaves, across induction and maintenance.
Objective response signal
A multimodal score tracks real change in mood-linked behavior rather than relying on the patient to report it.
Act before the no-show
Early signal gives your team time to reach out and keep the patient in treatment.
Privacy by design
Facial analysis runs on-device; voice is processed in a short ephemeral window and discarded, leaving only a numerical score.
The evidence
- Score validated against MADRS at r=0.89 across 11 studies.
- 75% activation when the physician recommends it, versus 30 to 40% for PHQ-9 email.
- No EHR integration needed; clinicians use a standalone dashboard.
Frequently asked questions
When does Spravato relapse usually begin?
Relapse risk is concentrated in the months after induction, and the earliest signals appear between REMS sessions. About half of patients relapse by six months without durable monitoring, and nearly 80% discontinue within a year (ESKALE).
Does the Spravato REMS cover between-session monitoring?
No. REMS governs supervised dosing, the minimum two-hour observation, and blood-pressure checks on dosing days. The days in between are not addressed, which is the window passive monitoring is designed to cover.
How is relapse detected without patient surveys?
Passive, multimodal monitoring measures facial, vocal, activity, and digital-behavior signals continuously, flagging deterioration roughly 48 hours before symptom-level worsening, with no action required from the patient.
Go deeper in the full guide: Spravato Outcomes Tracking: Clinical Guide
See it on a real patient case.
A 30-minute demo walks through the dashboard and how Spravato patients show up in the data between visits.