Remote Patient Monitoring for Critical Access Hospitals: A Program Your Existing Staff Can Run

Most critical access hospitals that won Rural Health Transformation Program awards in 2026 got money for remote patient monitoring without a care management department to run it. The award letter says, “remote monitoring for chronic disease.” The staffing chart says 25 beds, a rural health clinic, two or three nurses on a shift, and no one whose job is to look at blood pressure readings from 100 patients’ kitchens every morning.

That gap is the reason many hospital RPM programs stall between the award and the first enrolled patient. It is also solvable, because remote patient monitoring has two different models, and the difference between them is who watches the readings.

What does a critical access hospital need to run RPM?

Strip away vendor language and a remote monitoring program has four parts: 

• Devices that capture readings at home: a blood pressure cuff, a scale, a pulse oximeter, a glucose meter, matched to each patient’s conditions. For rural patients, the devices should connect over cellular networks, because a program that depends on home Wi-Fi or a smartphone will leave out the people who need it most.

• A platform that receives the readings, applies thresholds, raises alerts, and documents what happened.

• An EHR connection so readings and notes land in the patient’s chart instead of a separate login nobody opens.

• Clinical review by someone who looks at the alerts, calls the patient, and decides whether the provider needs to know. 

The first three are technology, and a vendor supplies them. The fourth is labor, and this is where the two models diverge.

Technology-only RPM: what your staff does 

In a technology-only program, the vendor provides the devices, logistics, platform, EHR integration, and training. Your nurses or medical assistants review the daily alert queue, call patients whose readings are out of range, document the contact, and route anything clinical to the provider.

What it asks of a 25-bed hospital’s staff, in practice:

• Alert review. A well-tuned program with 100 enrolled patients generates perhaps 10 to 20 alerts a day, most of them resolvable in a two-minute call. Plan for one person spending one to two hours each morning on the queue.
• Outreach for missed readings. Adherence drifts without a nudge. Budget another hour or two a week of phone calls.
• Monthly management time. Medicare’s management codes require 20 minutes of staff time per patient per month, cumulative, including the calls above. Across 100 patients, that is roughly 33 hours a month, which is close to a half-time position.
• Enrollment. Consent, device handoff, and a few minutes of instruction at a clinic visit or discharge. Front-office or discharge staff can absorb this.

Add it up and technology-only RPM for 100 patients is about 0.5 to 0.6 FTE of nursing or care coordination time. That is manageable if you have the person. It is not manageable if the plan is for the floor nurse to check the queue between admissions, because the queue will go unchecked on the first busy day and the program will quietly stop producing results.

Technology-only fits a CAH that already has a care coordinator, a chronic care management program, or a rural health clinic with a nurse who owns follow-up. It is the lower-cost option, and it keeps every patient contact inside your walls.

Clinically managed RPM: what the vendor’s nurses do 

In a clinically managed program, the vendor’s licensed nurses review readings daily, validate and triage alerts, call patients, provide education, and escalate to your providers under protocols your medical staff approves. Your providers keep every diagnostic and treatment decision. Your hospital’s role is referral, oversight, and the occasional visit that a reading moves up.

What it asks of your staff:

• A physician or nurse practitioner champion to approve thresholds and escalation rules at the start, and to respond to escalations, which in a stable program is a few per week.
• Front-office or discharge staff to enroll patients and hand out devices.
• A quarterly hour to review the program dashboard.

The clinical labor moves to the vendor, which is why this model costs more per patient per month, typically two to three times the technology-only rate. For a CAH with no care management staff, it is the model that gets patients enrolled before the first grant reporting deadline. Under the CY 2027 final rule, vendor staff time can support Medicare billing through December 31, 2028. From 2029, only your employed staff’s time counts. That makes clinically managed RPM a launch model for Medicare patients: run it while the program ramps, then move panels to your own staff before 2029. Patients can be assigned to either model, and the assignment can change month to month.

Which RPM model fits your hospital? 

In a clinically managed program, the vendor’s licensed nurses review readings daily, validate and triage alerts, call patients, provide education, and escalate to your providers under protocols your medical staff approves. Your providers keep every diagnostic and treatment decision. Your hospital’s role is referral, oversight, and the occasional visit that a reading moves up. 

What it asks of your staff: 

• A physician or nurse practitioner champion to approve thresholds and escalation rules at the start, and to respond to escalations, which in a stable program is a few per week. 

• Front-office or discharge staff to enroll patients and hand out devices. 

• A quarterly hour to review the program dashboard. 

The clinical labor moves to the vendor, which is why this model costs more per patient per month, typically two to three times the technology-only rate. For a CAH with no care management staff, it is the model that gets patients enrolled before the first grant reporting deadline. Under the CY 2027 final rule, vendor staff time can support Medicare billing through December 31, 2028. From 2029, only your employed staff’s time counts. That makes clinically managed RPM a launch model for Medicare patients: run it while the program ramps, then move panels to your own staff before 2029. Patients can be assigned to either model, and the assignment can change month to month. 

 Which RPM model fits your hospital?  

Ask three questions:

• Who will open the alert queue at 8 a.m. tomorrow? If you can name the person and their hours, technology-only is viable. If the answer is “we’ll figure it out,” start clinically managed.
• How many patients in year one? Under 50, a part-time coordinator can run it. Over 100, the management minutes alone justify either a dedicated position or the vendor’s team.
• What does the grant pay for? RHTP funds cover startup and implementation costs: devices, platform, integration, training, and time-limited staff. They do not cover services Medicare already reimburses. That matters for the next section.

Technology-only vs. clinically managed RPM (comparison table)

How RPM pays for itself after the RHTP grant 

The grant buys the infrastructure once. Reimbursement keeps the program running. Medicare pays for remote physiologic monitoring through a family of codes: 99453 for setup and patient education; 99454 for device supply when a patient transmits 16 or more days in a 30-day period, and 99445, new in 2026, for 2 to 15 days; 99457 for the first 20 minutes of monthly management, 99458 for each additional 20, and 99470, new in 2026, for 10 to 19 minutes. Chronic care management (99490) can be billed alongside where you provide it. 

Under either model, the management minutes are billable when they are documented. The difference is whose minutes they are. Under technology-only, they are your staff’s. Under a clinically managed program, through 2028, they are the vendor’s, under your providers’ general supervision and subject to your billing and compliance team’s review of how your hospital bills professional services. Starting January 1, 2029, only minutes from clinical staff your hospital employs count toward Medicare payment. The vendor’s platform should produce a monthly billing-support report showing reading days and minutes per patient, so claims are not reconstructed from memory. 

At 2026 national average rates, a Medicare patient who transmits 16 days and receives 20 minutes of management generated roughly $100 to $120 a month before chronic care management. The CY 2027 final rule revalued RPM services, so check 2027 rates in the Physician Fee Schedule lookup. Critical access hospital payment methods also differ, so confirm the figures against your own reimbursement. 

Two rules keep the grant and the reimbursement from colliding. Grant funds pay only for costs that are not reimbursable, and anything a payer will reimburse gets billed to that payer. Most state RHTP programs require a written attestation to that effect, and a vendor that has worked under 2 CFR Part 200 should be able to show you a cost allocation method before you sign. 

Starting without stalling

The hospitals that get RPM running within a quarter of the award tend to do four things: pick the model based on who will watch the queue, not on price; start with the patients whose readmissions hurt most, usually heart failure and hypertension, and add diabetes and COPD once the workflow holds; enroll at discharge, where a patient leaving with a kit and a 30-day plan is the fastest route to the readmission measure, and schedule the initiating visit the 2027 rule requires; and insist that EHR integration and billing setup are part of implementation rather than a later project.

None of this requires a care management department. It requires a decision about who watches, made before the devices ship.

AMC Health provides remote patient monitoring to critical access hospitals, FQHCs, and health systems under both models described here, including the VA Home Telehealth Program. This article is general information, not billing or legal advice; reimbursement varies by locality and payment method.