Pay for the decision

Medicare will stop paying for remote monitoring done by outside contractors in 2029, not 2027. Practices should use the time to make acting on the readings somebody's job.

Illustration of a bathroom scale with a small antenna, the light at its tip picked out in amber.

Medicare has given doctors who hand remote monitoring to outside companies two more years. It pays for the device that sends readings from a patient's home, a cuff or a scale, and for the staff who read them. In its final payment rule for 2027, released on Nov. 4, the agency kept its plan to pay only when clinical staff employed by the practice do that work, not contractors. But it pushed the start to Jan. 1, 2029, “to provide additional time for practices to adapt to this aspect of the policy.”

The rest arrives in January. Therapeutic monitoring, one of the two kinds Medicare pays for, will be limited to established patients, as physiologic monitoring already is. Whoever bills for either must provide a separate, billable visit when monitoring begins. Payment is being revalued because, the agency wrote in July, “the devices may be available at a reduced cost compared to our initial estimates.” On contractors, the agency wrote that outsourcing can fragment care and leave the billing clinician too little involved. Medicare should pay for readings only when changing the treatment is somebody's job, and practices should spend the two years building that job into their own care teams.

The hurry comes from the federal health department's inspector general. From 2019 to 2022, the number of Medicare patients on remote monitoring rose more than tenfold. In 2024 Medicare paid more than $500 million for it. Some of it may have bought nothing. The inspector general has warned of companies that call, text or advertise to Medicare patients and sign them up. “Most often, the monitoring never happens, but the company bills for remote patient monitoring anyway.” One provider billed 23,569 hours of treatment management in 2022. A year has 8,760 hours. Medicare could not tell, the inspector general wrote, whether that was unreasonable, because it did not know how many people did the work.

Monitoring helps when the readings change the treatment, and the clearest trials are in blood pressure, the most common reason Medicare patients were monitored. In a Minnesota trial of 450 adults with uncontrolled pressure, half sent home readings to pharmacists, who adjusted their medicines. At six months and at a year, 57 percent of them had it under control, nearly twice the share with usual care. In British general practices, when doctors adjusted drugs using patients' home readings rather than clinic readings, systolic pressure a year later was three to five points lower. In both trials somebody changed a dose. That was the treatment.

Heart failure is harder. In one American trial, patients recently hospitalized for heart failure reported their weight and symptoms each day to an automated phone line, and their own clinicians reviewed the reports. About half in each group were readmitted or died within six months, monitored or not. In another, in California, nurses at a central site reviewed the readings, followed protocols and phoned patients. Readmissions did not fall. A German trial, paid for by its government, tested a program meant to enable “a prompt initiation of the appropriate treatment and care.” Its patients lost about 18 days a year to unplanned cardiovascular admissions or death, against 24 with usual care, and fewer died of any cause.

The German patients kept their usual doctors, but their readings went to a central telemedical center at a Berlin university hospital. Heart failure nurses and physicians there could change a medicine or send a patient to an emergency room, and a physician took emergency calls around the clock. In the three trials that worked, changing the treatment was built into the program. In the two that did not, the patients' own clinicians or central nurses following protocols reviewed the readings, and the published summaries do not say how often anything changed. None of the five compared employed staff with contractors, or a fast response with a slow one.

Vendors and their clients have a case. The Medical Group Management Association told Medicare that vendors offer “economies of scale” and that some practices with staff “already stretched thin” could not bring the work in-house. A coalition of more than 200 organizations warned of disruption for about 1 million Medicare patients and said “Small practices, rural providers, and safety-net organizations would be hit hardest, where patients often face long travel distances, fewer clinicians and specialists, hospital closures, and limited access to care.” The American Telemedicine Association's chief called cutting fraud “a goal we share,” but not at patients' expense. They wrote before the delay. The association's chief has since welcomed it, but said that “a delay is not a fix.” They are probably right about who would be hit, and the German center suggests that a team outside the practice can do the job. Medicare's rule on contractors asks only who employs the staff, though the agency has promised guidance before 2029 on which staffing arrangements may be appropriate. The trials suggest it should ask whose job it is to change the treatment.

Practices should start now. A small or rural practice does not need a call center. It needs the readings in its own record, in front of a nurse or pharmacist with standing orders to adjust a blood pressure drug or a diuretic. It should enroll the patients whose treatment the numbers will change. Federal rural transformation money can pay for training rural hospitals in remote monitoring. Medicare pays today by the days a device sends data and the minutes of contact. The four new codes it asked about should pay for a reading that led to a decision.

Medicare has given practices two years, and the inspector general has given them the reason. Its 2024 report describes patients signed up by a phone call or a text, for monitoring that most often never happened. The trials show the alternative. Someone whose job was to change the treatment saw the numbers, and changed it.

  • What the final rule did
    CMS's fact sheet on the calendar year 2027 Medicare Physician Fee Schedule final rule, dated Nov. 4, 2026, lists four remote monitoring policies it had proposed: remote therapeutic monitoring (RTM) furnished only to established patients; a separately reportable initiating visit for practitioners reporting remote physiologic monitoring (RPM) or RTM; updates to how the services are valued; and payment only when the services are performed by clinical staff employed by the practice and not by contractors. CMS finalized the first three for 2027, except a proposal to remove clinical labor costs from six treatment management codes. It delayed the fourth for two years, to Jan. 1, 2029, to give practices additional time to adapt; the requirement applies to RPM and RTM services beginning with CY 2029. CMS also sought comment on bundling the RPM and RTM CPT codes and creating four new HCPCS G-codes, and says it may consider the comments in future rulemaking. The 2026 rule had added codes for fewer than 16 days of data transmission per 30-day period and fewer than 20 minutes of interactive communication per month.
  • What CMS said when it proposed the changes
    CMS's fact sheet on the proposed rule, July 14, 2026, says the agency proposed updating how remote monitoring is valued because it understands the devices may be available at a reduced cost compared to its initial estimates, and that the bundled G-code approach on which it sought comment would address recommendations from recent OIG reports that it does not believe can be fully resolved with the current coding structure. A July 28, 2026 alert by the law firm Benesch attributes to CMS the concern that outsourced monitoring can fragment care and lead to insufficient involvement and oversight of the billing practitioner. The Medical Group Management Association's comment letter says the established patient requirement has been in place for RPM since the end of the COVID-19 public health emergency.
  • The inspector general's 2024 report
    Additional Oversight of Remote Patient Monitoring in Medicare Is Needed (OEI-02-23-00260, September 2024). In 2022 slightly more than 570,000 enrollees received remote patient monitoring, compared with about 55,000 in 2019, and payments were more than $300 million, compared with $15 million in 2019. More than half of enrollees received it for hypertension. About 43 percent did not receive all 3 components; about 28 percent had no claim or encounter record for education and setup, and 12 percent did not receive treatment management. A single provider billed 23,569 hours of treatment management in 2022; because CMS does not know how many individuals delivered the services, OIG wrote, it cannot determine whether that is an unreasonable number of hours. About 1 in 10 enrollees received monitoring from a company that appeared to specialize in it. OIG recommended, among other steps, that CMS require monitoring to be ordered and identify and monitor companies that bill for it.
  • The inspector general's 2025 report and its warning about sign-ups
    Billing for Remote Patient Monitoring in Medicare (OEI-02-23-00261, issued Aug. 25, 2025) found that use continued to grow in 2024, with Medicare payments exceeding $500 million. It set out billing indicators that can identify practices warranting further scrutiny, among them billing for a high proportion of enrollees with no prior history with the practice and billing for multiple devices a month for one enrollee. The 2024 report recounts OIG's November 2023 consumer alert about companies that make unsolicited contact with enrollees by calls, texts or internet ads and sign them up for remote patient monitoring, and says that most often the monitoring never happens but the company bills anyway.
  • The blood pressure trials
    Margolis and colleagues (JAMA, 2013) ran a cluster randomized trial in 16 primary care clinics of an integrated health system in Minneapolis-St Paul: 450 adults with uncontrolled blood pressure, 228 in clinics offering home telemonitoring with pharmacists who adjusted therapy, and 222 in usual care. Blood pressure was controlled at both 6 and 12 months in 57.2% versus 30.0%. McManus and colleagues (TASMINH4, Lancet, 2018) randomized 1,182 patients in 142 UK general practices to self-monitoring, self-monitoring with telemonitoring, or usual care, with general practitioners titrating medication. At 12 months systolic pressure was lower than with usual care by 3.5 mm Hg with self-monitoring and 4.7 mm Hg with telemonitoring; the two intervention groups did not differ. Its funders included the National Institute for Health Research and Omron Healthcare UK.
  • The heart failure trials
    Tele-HF (NEJM, 2010; funded by the National Heart, Lung, and Blood Institute) randomized 1,653 patients recently hospitalized for heart failure to telephone-based reporting of symptoms and weight, reviewed by their clinicians, or usual care; readmission or death within 180 days occurred in 52.3% versus 51.5%. BEAT-HF (JAMA Internal Medicine, 2016) randomized 1,437 patients in California to coaching calls plus telemonitoring reviewed by centralized registered nurses, or usual care; 180-day readmission was 50.8% versus 49.2%. TIM-HF2 (Lancet, 2018; German research ministry), unmasked, with 1,571 patients, found 4.88% versus 6.64% of days lost to unplanned cardiovascular admission or death (ratio 0.80, 95% CI 0.65 to 1.00; 17.8 versus 24.2 days a year) and all-cause death of 7.86 versus 11.34 per 100 person-years (hazard ratio 0.70); cardiovascular deaths did not differ significantly. Its registry entry says participants would continue usual care from their general practitioner and specialist. The trial team's 2021 paper describes a dedicated center at the Charité in Berlin, staffed by heart failure nurses and physicians, with a physician-led emergency service 24/7 that could change medication or send a patient to an emergency room.
  • What practices and hospitals told CMS, and the rural money
    The Medical Group Management Association's comment letter of Sept. 14, 2026, signed by Anders Gilberg, urged CMS to withdraw the remote monitoring proposals and said many medical groups partner with third-party vendors who provide the services through economies of scale. A coalition letter to Dr. Oz dated Aug. 24, 2026 urged CMS to delay the policies, said they would disrupt care for approximately 1 million Medicare beneficiaries and that small practices, rural providers and safety-net organizations would be hit hardest. Becker's reported more than 200 signers, including more than 30 health systems and hospitals. A Texas Health and Human Services Commission briefing of Feb. 11, 2026 lists among the Rural Health Transformation Program's permitted uses training and technical assistance to improve care delivery in rural hospitals, including remote monitoring, and lists active remote monitoring for high acuity patients as one Texas option.
  • What the telemedicine industry said
    In a July 21, 2026 release from ATA Action, the American Telemedicine Association's affiliated advocacy organization, Kyle Zebley, chief executive of the association and executive director of ATA Action, said the proposed rule includes troublesome provisions scaling back third-party remote physiologic and therapeutic monitoring and cuts to their reimbursement. He said CMS is understandably looking to reduce fraud, waste and abuse, a goal the group shares, but that it should not come at the expense of providing remote monitoring for patients and health care providers. In a Nov. 5, 2026 release welcoming the delay, Zebley said that CMS listened, that a delay is not a fix, and that ATA Action will use the next two rulemaking cycles to press CMS to adopt targeted program-integrity safeguards that address fraud and abuse without cutting off access to proven remote monitoring programs.

Latest news

  1. Medicare Advantage plans would have to make in-network clinicians file prior authorizations through an electronic interface, CMS proposes
  2. Medicare's first mandatory Advantage model would swap star ratings for local 'Diamond Ratings' in randomly chosen states from 2029
  3. A California health system says an employee uploaded emergency patients' diagnoses and record numbers to ChatGPT

See all the latest news