Remote Therapeutic Monitoring
The agent runs the recovery conversation.
We turn it into billed, coded, signed revenue.
The market is filling up with patient-facing AI that monitors post-op recovery and PT adherence. That is the conversation layer, and it is getting good. But Medicare does not pay a conversation. It pays a named, enrolled clinician who furnishes the monitoring, reviews it, and signs the treatment-management decision.
SurgeonValue is that rail. Bring your own agent. We turn what it hears into enrolled, coded, physician-signed, audit-survivable RTM.
ABOVE — the agent
A patient-facing AI agent runs the conversation.
Post-op check-ins at 2am. PT adherence nudges. “Is this swelling normal?” It is genuinely good at the conversation, and it generates a stream of monitoring data. Bring whichever one you already use — this rail sits under any of them.
THE GAP — nobody is on the record
That monitoring is RTM-billable. An agent cannot bill it.
Medicare pays remote therapeutic monitoring only when a named, enrolled clinician furnishes it, reviews the data, and stands behind the treatment-management decision. A patient never enrolled. The device-supply and management codes were never captured. No clinician reviewed or signed. The monitoring happened; the revenue — and the accountability — fell on the floor.
UNDER — SurgeonValue, the attested rail
We turn agent-run monitoring into enrolled, coded, signed RTM.
Eligibility and enrollment detected from the panel. The RTM codes captured and checked against the CMS NCCI engine — the model proposes, the engine grades, the model never grades its own work. The treatment-management decision put in front of the surgeon, who reviews and signs. A hash-anchored receipt — who furnished it, who signed, when — that survives an audit. Your practice bills; we are the rail.
The codes that fall on the floor
For a monitored orthopedic recovery, the RTM stack is concrete. None of it bills without an enrolled clinician who furnishes and signs — which is exactly the part an autonomous agent cannot be.
Codes reflect 2026 Medicare RTM (MSK). Enrollment, documentation, and the clinician-review requirement gate every one — your practice bills them, not us. Illustrative, not coding advice.
Four rules that decide whether the maths works
Most RTM models we see are wrong before they start, because they miss one of these. All four are Medicare requirements, not opinions.
1 · Not during your own global period
“For global periods of surgery, remote physiologic monitoring and RTM may be billed by practitioners that are notreceiving the global service payment.” The operating surgeon is receiving it. So on your own surgical patient, your RTM revenue does not begin at discharge — it begins after the global period ends. Inside that window the perimeter is paid for through the episode, not through monitoring.
2 · One practitioner per patient per 30 days
The monthly slot is exclusive. Whoever enrols the patient holds the month, and a physical therapist or another practitioner can hold it instead of you. Monitoring is not a service line you add later at your convenience; it is a slot somebody occupies.
3 · RTM needs no established patient relationship
Remote physiologic monitoring requires one. RTM does not. That single asymmetry is why the scalable population for a surgeon is not only the post-operative panel.
4 · General supervision — this is the one that scales
“The services may be provided by health care personnel under the general supervision of the billing practitioner.” General, not direct. The practitioner need not be present. This is the only line in the rules that lets one clinician’s accountability cover a population rather than a schedule — and it is the reason a monitoring panel is not capped by clinic hours.
Source: telehealth.hhs.gov, “Billing for remote patient monitoring”, page last updated 17 January 2025 and verified live 10 September 2026. Coding guidance changes; confirm against the current Physician Fee Schedule before you build a model on it. Payment amounts vary by locality and year, and nothing here is billing advice.
Who signs the 98%?
The agent platforms advertise that most patient interactions never need a human. That is a feature — until a regulator, a payer, or a plaintiff asks who was accountable for the ninety-eight percent. A growing list of states now requires a licensed human to make the medical-necessity decision, and Medicare's remote-monitoring rules keep tightening around who is actually allowed to furnish and bill the service.
An autonomous agent can hold the conversation. It cannot be the name on the record. SurgeonValue puts a licensed physician's signature on every billable determination and seals it into a receipt anyone can verify — the one thing the conversation layer structurally cannot provide.
See the rail actually run
The attested loop →
Read a signed note, the model proposes, the NCCI engine checks, a clinician signs, a hash-anchored receipt goes back to the chart. Live sandbox, no PHI.
Attested PROMs → $0/collection
A validated outcome with a signature and a receipt — the measure value-based care pays on, captured alongside the monitoring.
SurgeonValue does not furnish monitoring or bill Medicare. It is the tooling a surgeon's own practice uses to capture, check, and attest the care it already delivers. Agent platforms named nowhere here; this rail is vendor-neutral by design.