Connected medical devices are no longer a pilot project. Infusion pumps, patient monitors, imaging systems, telehealth carts, remote monitoring kits, and tracked assets all sit on the network now, and most of them sit there permanently. Claroty’s Team82 analyzed 2.25 million IoMT devices across 351 healthcare organizations in its 2025 healthcare exposures report, a fleet size that would have been unimaginable a decade ago.
The hard part is no longer getting a device online. The hard part is keeping thousands of devices online across buildings, campuses, mobile clinics, and patient homes, on networks the health system does not own, and proving it stayed that way.
Most connectivity guidance for healthcare stops at a technology comparison. Cellular beats Wi-Fi, Wi-Fi beats Bluetooth, end of article. That framing misses what actually breaks in production. Connectivity failures in healthcare are rarely a radio problem. They are a coverage problem, a carrier problem, and above all a management problem.
What is healthcare IoT connectivity?
Healthcare IoT connectivity is the set of wireless networks, SIM technologies, and management systems that keep connected medical devices, clinical equipment, and remote monitoring hardware transmitting data reliably and securely. It spans the radio layer, meaning cellular, Wi-Fi, Bluetooth, and satellite, the identity layer, meaning the SIM and the carrier profiles it carries, and the management layer, meaning the platform where connectivity is provisioned, monitored, and controlled.
Health systems that evaluate only the radio layer end up solving one third of the problem.
Why Wi-Fi is not a connectivity strategy for medical devices
Wi-Fi is the default because it is already there. It is also the layer with the least predictable behavior for medical devices, for three reasons.
Coexistence degrades performance measurably. Researchers at FDA’s Center for Devices and Radiological Health and NIST ran a wireless coexistence study using a Bluetooth Low Energy evaluation board as a stand-in for a wireless medical device, tested against Wi-Fi traffic in the same 2.4 GHz band. Baseline throughput of 340 Kbps fell to 246 Kbps with one active 20 MHz Wi-Fi channel, 185 Kbps with two, and 100 Kbps with three. That is a 71 percent loss under a controlled worst case of fully saturated Wi-Fi channels. Real hospital environments are less severe than the lab condition, which is exactly why the result matters. The degradation curve is continuous, and a busy floor sits somewhere on it. FDA names coexistence among the core risk considerations for wireless medical devices, alongside quality of service, security, and electromagnetic compatibility.
Coverage stops at the building. A patient monitor that works in a med-surg unit does not work in the parking garage, the ambulance bay, or the patient’s apartment. Remote patient monitoring in particular assumes connectivity in an environment no hospital IT team controls or can troubleshoot, and RPM is no longer a marginal program. HHS Office of Inspector General reported that Medicare payments for remote patient monitoring grew from $15 million in 2019 to $536 million in 2024, with nearly one million Medicare enrollees receiving RPM services that year.
Shared infrastructure means shared outages. When the enterprise network goes down, clinical devices go down with it. The July 2024 CrowdStrike outage produced detectable service disruptions at 759 of 2,232 US hospitals, roughly 34 percent, according to a study published in JAMA Network Open. Of the 1,098 distinct services affected, 239 were patient-facing. Among services with measurable recovery times, 43 were still down more than 48 hours later. The authors note their method captures internet-facing services only, making the figure a lower bound. ECRI ranked unpreparedness for a “digital darkness” event as the number two health technology hazard for 2026.
Cellular does not eliminate these risks. It moves critical devices off the shared enterprise path and onto an independent one, which is a meaningful change in blast radius.
The problem single-carrier cellular does not solve
Moving devices to cellular introduces a different dependency. A device provisioned with a traditional SIM carries one IMSI tied to one operator. That device can only ever see one network.
For a health system, three consequences follow.
Coverage is decided at procurement. A carrier that performs well at the flagship campus may perform poorly at a rural clinic, inside a basement imaging suite, or at a patient’s home. That variance is discovered after deployment, and the only remedy is a physical SIM swap on every affected device.
Sunsets are scheduled failures. Network retirements are not a one-time event that ended with 3G. AT&T retired 3G on February 22, 2022. Sprint’s CDMA network ended March 31, 2022, and T-Mobile’s 3G network followed on July 1, 2022. Verizon retired CDMA on December 31, 2022. T-Mobile retired its 2G GSM network on August 3, 2026. AT&T announced in November 2024 that it would stop NB-IoT sales and certifications and decommission the network, targeting the first quarter of 2025, with customers migrating to LTE-M. That last one is worth noting because it happened at the radio-technology layer without any “shutdown” headline at all.
The 3G transition showed what this looks like in healthcare specifically. Roughly two million security, fire, and medical alert devices were still on 3G as AT&T’s shutdown approached, including hundreds of thousands of personal emergency response systems, according to an Alarm Industry Communications Committee survey. AICC petitioned the FCC to delay the shutdown, and AARP was among the groups backing the request, arguing that devices seniors relied on would stop sending alerts. Medical alert providers reported spending hundreds of thousands of dollars on upgrades and customer service to manage the transition.
One vendor relationship becomes many. Health systems that solve coverage variance by adding carriers inherit a second problem. Each carrier means a separate contract, a separate SIM SKU to stock, a separate portal to check, and a separate bill to reconcile. The fleet gets more reliable and the operation gets less manageable.
Why multi-carrier connectivity works for devices in the field
A multi-carrier SIM gives a device access to more than one cellular network from a single SIM installation. Instead of being tied to whichever carrier was chosen at purchase, the device can use whichever available network actually performs where it happens to be.
Three things follow from that, and each maps to a problem health systems already have.
The device adapts to its location instead of the other way around. Coverage is not uniform, and it is not uniform in ways that are difficult to predict from a coverage map. A carrier that is strong across a metro campus may be weak inside a particular basement, at a clinic forty miles out, or along a specific stretch of highway. A device with access to multiple networks finds the one that works at that spot, without anyone deciding in advance which carrier each device should get.
Connectivity survives problems on any one network. Congestion, a tower outage, a regional service issue, or an eventual network retirement takes down every device locked to that carrier at the same time. A device that can reach another network keeps reporting. For equipment that is monitoring a patient or holding a cold chain, the difference between degraded and offline is the whole point.
Networks can change without touching the hardware. Carrier assignments on a multi-carrier SIM are updated remotely, over the air. No one opens an enclosure, visits a site, or ships a replacement. That matters most for the devices that are hardest to reach, including sealed and patient-attached equipment, devices deployed to patient homes, and anything mounted in a vehicle or installed at a site with no on-site IT staff.
The last point is what makes this a lifecycle decision rather than a procurement detail. Medical devices routinely stay in service for seven to ten years. The networks under them do not stay still that long, as the 3G and 2G retirements demonstrated. A fleet that can change carriers remotely absorbs those changes as a configuration update. A fleet that cannot absorbs them as a hardware refresh.
The operational benefit is simpler still. One SIM reaching multiple carriers means one part to source and stock, one relationship to manage, and one set of terms, rather than a separate SIM, contract, and portal for every network the deployment touches.
Comparing connectivity options for healthcare devices
| Option | Coverage | Reliability profile | Fleet management | Best fit |
| Wi-Fi | Inside facility only | Subject to coexistence interference and shared enterprise outages | Managed by hospital IT alongside all other traffic | Non-critical in-facility devices, high-bandwidth imaging transfer |
| Bluetooth / BLE | Meters | Requires a gateway or paired phone to reach the network | Dependent on the gateway device | Wearables and sensors paired to a hub |
| Single-carrier cellular | Wherever that one carrier reaches | Fails wholesale when that carrier degrades or sunsets | One portal, one contract, no alternative network | Single-site deployments in strong coverage areas |
| Multi-carrier cellular (multi-IMSI + eUICC) | Any network with a profile on the SIM | Device shifts carriers autonomously; profiles updated over the air | One SIM, one relationship, one platform | Distributed fleets, RPM, mobile and rural care, long-lifecycle devices |
| Satellite (Starlink) | Beyond terrestrial coverage | Independent of cellular infrastructure | Managed separately unless consolidated | Rural sites, mobile clinics, disaster response, failover for critical locations |
The layer most connectivity guides skip
A health system running 4,000 connected devices across 12 sites does not experience connectivity as a technology. It experiences connectivity as an operational surface. Which devices are online. Which SIM belongs to which asset. Which device burned through its data plan this month and why. Which sites saw network events last night. Whether the bill matches the deployment.
That information lives in carrier portals by default, one portal per carrier, with no shared view and no shared vocabulary. Consolidating carriers into a single SIM solves the coverage problem. It does not by itself solve the visibility problem.
This is the distinction between the connectivity layer and the management layer, and it is worth keeping separate when evaluating vendors.
The DB1 SIM is the connectivity layer. It provides access to multiple carriers through a single SIM installation, deployable as a standard physical SIM in 2FF, 3FF, or 4FF form factors, or as an embedded SIM. One SIM, multiple carriers, no per-network sourcing and stocking.
VOYAGER is the management layer. It is where those SIMs, IMSIs, data plans, usage patterns, and network events are monitored and controlled, alongside the endpoints themselves. Cellular and Starlink connectivity are managed in the same platform rather than in separate systems.
The DB1 keeps devices connected across carriers. VOYAGER gives the team visibility and control over that connectivity. Health systems evaluating connectivity vendors should confirm both layers are addressed, because a multi-carrier SIM managed through six disconnected dashboards recreates the problem it was bought to solve.
How connected devices should reach the network
Getting a device onto a carrier network is the first decision. Deciding where that traffic goes next is the one that matters for security reviews.
Private APNs keep device traffic off the public internet. An APN, or Access Point Name, is the gateway a cellular device uses to reach a data network. A public APN drops the device onto the open internet with a dynamic address. A private APN routes traffic into a defined, closed path instead, so devices are not internet-reachable and are not discoverable by anything scanning public address space. For a fleet of clinical devices, this converts thousands of individually exposed endpoints into a single controlled ingress point.
VPN tunneling connects that path to the health system. A private APN paired with an IPsec tunnel or a direct interconnect delivers device traffic into the organization’s own network, encrypted end to end, without traversing the public internet at any point. Static or fixed IP addressing on top of that allows firewall rules, allow lists, and network segmentation to be written against known device addresses rather than whatever the carrier assigns that day. Segmentation matters here in particular, because clinical devices frequently cannot be patched on an IT schedule and need to be isolated rather than hardened.
Centralized management is what makes the architecture hold. Private APNs, tunnels, and IP assignments are straightforward to design and difficult to maintain across multiple carriers, because each carrier implements and exposes them differently. Managing three carriers means three sets of APN configurations, three addressing schemes, and three portals to check when a device stops reporting. A multi-carrier SIM managed in one platform collapses that into a single configuration and a single view. Provisioning, IP assignment, usage policy, and network events apply across the fleet rather than per carrier.
That consolidation also produces the operational record security and compliance teams ask for. Which devices are active, which networks they used, what they transmitted, and when connectivity changed are questions with one answer instead of several. Specific requirements vary by device class and by jurisdiction, but they point in the same direction. Device traffic should be contained, encrypted, segmented, and observable from one place. An architecture built that way answers most of the questions a security review will ask before it asks them.
An evaluation checklist for health system IT
Questions worth asking any connectivity vendor before signing.
- Can a single SIM reach more than one carrier, and does switching happen autonomously on the device or require a provisioning call?
- Is the same connectivity model available as both a physical SIM and an eSIM, so device design does not dictate connectivity strategy?
- Can carrier profiles be added or changed over the air, without physical access to deployed devices?
- Does one platform show SIMs, endpoints, usage, and network events together, or do those live in separate systems?
- Are usage alerts and data pooling configurable before an overage appears on a bill rather than after?
- Is satellite connectivity manageable in the same platform as cellular, or does it require a parallel operation?
- What is the escalation path when a device at a rural site drops off, and who owns the carrier relationship during that escalation?
Frequently asked questions
For devices that must stay online outside a single building, multi-carrier cellular is the strongest option. It provides coverage independent of hospital infrastructure, and unlike single-carrier cellular it does not fail wholesale when one network degrades or retires. Wi-Fi remains appropriate for high-bandwidth, in-facility, non-critical transfers.
An APN is the gateway a cellular device uses to reach a data network. A public APN places the device on the open internet. A private APN routes its traffic into a closed, defined path instead, so the device is not internet-reachable or discoverable by public scanning. Paired with VPN tunneling and fixed IP addressing, it lets an organization apply its own firewall rules and network segmentation to devices that often cannot be patched on an IT schedule.
No network is HIPAA compliant on its own. HIPAA compliance is a property of the covered entity’s overall safeguards. Cellular networks provide carrier-grade encryption over the air, and private APNs plus VPN tunneling keep device traffic off the public internet, which supports the transmission security standard at 45 CFR 164.312(e). Compliance still depends on the full implementation, including business associate agreements.
A multi-carrier SIM is a single SIM that gives a device access to more than one cellular network. Rather than being locked to the carrier chosen at purchase, the device can use whichever available network performs best where it is deployed, and its carrier assignments can be updated remotely without physical access to the hardware.
Devices on a single-carrier SIM stop working and require physical replacement or a SIM swap. Ahead of the 3G sunsets of 2022, roughly two million alarm and medical alert devices were still on 3G, and industry groups petitioned the FCC for a delay. Devices on a multi-carrier SIM with eUICC can be moved to another network profile remotely.
It depends on coverage, and specifically on where the devices will actually operate. A deployment contained to one well-covered campus may only need one carrier. A deployment that spans rural clinics, patient homes, vehicles, or multiple regions will hit places where the primary carrier is weak, and those gaps are difficult to predict from a coverage map before devices are in the field. Mapping the deployment footprint against real coverage in those locations is what determines the answer. The added benefit of a multi-carrier SIM is that the decision does not have to be made up front or permanently. Multiple carriers are available through a single SIM and a single relationship, so coverage can expand as the deployment does without adding a SIM, a contract, and a portal for every network.
Connectivity that scales with the deployment
Healthcare is one of the least forgiving environments for connectivity. Devices last longer than the networks they were designed for. Coverage requirements extend past the campus to mobile units, satellite clinics, and patient homes. Failures have clinical consequences, and evidence of reliability has regulatory ones.
Single-carrier connectivity, and the single carrier portal that comes with it, works until the deployment grows. Multi-carrier connectivity managed in one platform is what holds up when it does.
Talk with a Datablaze expert about connectivity for connected health deployments. Get started.
Sources
- Claroty Team82, State of CPS Security: Healthcare Exposures 2025, March 26, 2025. https://claroty.com/blog/clarotys-state-of-cps-security-report-healthcare-exposures-2025
- Al Kalaa, Guag, Seidman (FDA/CDRH); Ma, Coder (NIST), IEEE Electromagnetic Compatibility Magazine, Vol. 9, Q4 2020. https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=930144
- US FDA, Wireless Medical Devices. https://www.fda.gov/medical-devices/digital-health-center-excellence/wireless-medical-devices
- Tully et al., “Patient Care Technology Disruptions Associated with the CrowdStrike Outage,” JAMA Network Open, July 19, 2025. https://cyberhealth.ucsd.edu/_files/Patient%20Care%20Technology%20Disruptions%20Associated%20with%20the%20CrowdStrike%20Outage.pdf
- ECRI, Top 10 Health Technology Hazards for 2026. https://assets.ecri.org/PDF/ECRI_Top_10_Health_Technology_Hazards_for_2026_Executive_Brief.pdf
- T-Mobile, Network Evolution (2G GSM retirement August 3, 2026; 3G UMTS July 1, 2022; Sprint CDMA March 31, 2022). https://www.t-mobile.com/support/coverage/t-mobile-network-evolution
- FCC, Plan Ahead for Phase-Out of 3G Cellular Networks and Service. https://www.fcc.gov/consumers/guides/plan-ahead-phase-out-3g-cellular-networks-and-service
- Verizon Business, CDMA Network Retirement. https://www.verizon.com/business/support/services-and-apps/cdma-network-retire/
- Light Reading, “AT&T to discontinue NB-IoT,” November 20, 2024. https://www.lightreading.com/iot/at-t-to-discontinue-nb-iot-but-t-mobile-and-verizon-keep-the-faith
- CNBC, “AT&T 3G shutdown to impact seniors with medical alert devices,” February 19, 2022. https://www.cnbc.com/2022/02/19/att-3g-shutdown-on-feb-22-to-impact-seniors-with-medical-alert-devices.html
- IEEE Spectrum, “AT&T’s 3G Shutdown,” March 1, 2022. https://spectrum.ieee.org/att-3g-shutdown
- HHS Office of Inspector General, Billing for Remote Patient Monitoring in Medicare, OEI-02-23-00261, August 25, 2025. https://oig.hhs.gov/reports/all/2025/billing-for-remote-patient-monitoring/
