The remote patient monitoring reset
The first generation of RPM was a reimbursement play. The second generation is a clinical outcomes play — and it looks very different.
What broke
RPM programs built around CPT 99453/99454 billing hit their ceiling. Payers tightened medical necessity requirements, patient engagement decayed after 90 days, and the unit economics stopped working for many operators.
What replaces it
Second-generation RPM is being built inside condition-specific care programs — heart failure, CKD, high-risk pregnancy — where the device is a means, not the product. The winning teams sell to at-risk providers, not fee-for-service billers.
Where founders should focus
Own the clinical protocol, not the sensor. The sensor commoditizes. The protocol, the escalation pathway, and the integration into the medical group's workflow are the durable moats.
The billing-code arbitrage collapses
The first RPM wave was built substantially around a small set of monthly billing codes tied to device data transmission thresholds, and once payers tightened documentation and medical-necessity requirements around those codes, a meaningful share of RPM revenue simply evaporated for companies that had optimized for code capture rather than clinical value. Programs that could not point to a change in a clinical outcome measure struggled to defend their billing in audits.
This was a predictable outcome of building a business model around a reimbursement mechanic rather than a clinical thesis, but it took actual payer scrutiny to force the correction. Companies that survived the reset were disproportionately the ones that had, even before the tightening, been tracking hospitalization or ED utilization changes alongside billing metrics.
Device sprawl versus workflow depth
A second-generation RPM program increasingly looks narrower in device scope but deeper in clinical workflow — fewer conditions monitored per patient, but tighter integration with care management staffing, escalation protocols, and medication titration decisions. The earlier model of shipping as many device types as possible to as many patients as possible optimized for enrollment metrics that looked good in a sales deck but did not correlate well with outcomes.
Programs now succeeding tend to pair monitoring with a defined clinical action pathway: a specific threshold triggers a specific intervention by a specific role within a specific time window. Without that operational scaffolding, monitoring data simply accumulates unread, which was in fact the dominant failure mode of the first generation.
The staffing model founders underestimate
Remote monitoring only produces outcomes if someone acts on the signal, and that someone is usually a care manager whose caseload economics are unforgiving. A program that monitors too many patients per care manager degrades into alert fatigue and missed escalations, while a program with a conservative caseload struggles to justify its cost per patient to a finance team.
The companies getting this right are treating care manager staffing ratios as a first-class product design variable, not an implementation afterthought, and are building software that actively manages the manager's queue by clinical priority rather than by raw alert volume. That queue-management layer is now arguably more valuable, and more defensible, than the monitoring device itself.
Where the founder opportunity actually sits
Founders entering this space now face a market that already has device supply solved; the open problem is the orchestration layer that turns a stream of vitals into a prioritized, staffed clinical action. That is a harder, less venture-familiar problem than shipping a Bluetooth blood pressure cuff, but it is also far more defensible, since it requires deep integration into a health system's staffing and escalation workflows rather than a one-time procurement decision.
A useful discipline for new entrants is to design the staffing model and the reimbursement model together from day one, rather than assuming a billing code will make the staffing economics work out. Programs built that way have a much easier time surviving the next round of payer scrutiny, whenever it comes.
