11 March 2026 · 12 min read · Arviteni
Care homes need environment monitoring, falls detection, and biomarker tracking. Most solutions lock you into proprietary hardware. This guide covers the sensor landscape and what to look for.
Care homes are responsible for monitoring a lot at once: room temperatures for medication safety, occupancy patterns for night checks, falls risk for every high-dependency resident, and increasingly, biomarkers like blood pressure and oxygen saturation. Most providers do this manually, or with a collection of disconnected systems that do not talk to each other.
The technology to do this well exists. The sensors are mature, reliable, and increasingly affordable. The problem is not the hardware. The problem is how the market sells it.
Most monitoring vendors sell proprietary ecosystems. Their sensors only work with their platform. Their platform only exports to formats they control. If you want to switch provider, you replace every sensor in the building. That dependency is not accidental. It is a business model. This guide explains what is actually happening at the technology level, what good sensor infrastructure looks like, and how to evaluate platforms before you commit.
Before discussing protocols and hardware, it is worth being precise about what you are trying to achieve. There are four distinct monitoring categories, and they require different sensor types.
Environmental monitoring covers room temperature, humidity, and air quality. For care homes, the most pressing use case is medication storage. Most medicines have defined temperature ranges, and CQC inspectors increasingly expect evidence that storage conditions are being tracked, not just checked manually at spot intervals. Beyond medication, high temperatures correlate with increased falls risk and dehydration, and humidity affects infection control.
Occupancy and motion monitoring tells you whether a room is occupied and, broadly, what level of activity is occurring. This supports night checks, flagging when a resident has not moved for an unusually long period, and generating evidence for staffing decisions.
Falls detection is the highest-stakes category. A resident who falls and is not found quickly faces serious harm. Conventional approaches rely on residents pressing a pendant alarm, which assumes the resident is conscious, mobile enough to press it, and actually wearing it. All three assumptions frequently fail.
Biomarker monitoring covers vital signs: blood pressure, weight, SpO2 (blood oxygen saturation), heart rate, and glucose for diabetic residents. This is typically done manually during care rounds, but connected devices can automate the recording and flag readings that fall outside normal ranges before they become a clinical incident.
Each category has specific hardware requirements. A temperature sensor suitable for a medication cupboard is not the same device as a radar-based falls detector. A platform that handles all four categories well is harder to find than one that does two of them adequately.
When you speak to monitoring vendors, you will encounter protocol names. Understanding what these mean helps you evaluate whether a product is genuinely flexible or whether it is locked into a single approach.
Zigbee is a mesh radio protocol that has been around since 2003 and is the most widely deployed standard in building sensor networks. Devices communicate at 2.4 GHz and form a mesh: each device can relay traffic for neighbouring devices, extending coverage without needing every node to reach a central hub directly. Power consumption is very low. Battery-powered Zigbee sensors routinely run for two to three years on a single battery, which matters when you have sensors in dozens of rooms. Range between nodes is typically 10 to 30 metres indoors, but because of the mesh topology, you rarely need to worry about distance from the hub. The limitation is that Zigbee is not IP-native: you need a Zigbee coordinator (a small hub device) to translate between the Zigbee mesh and your IP network.
Thread is a newer mesh protocol, released in 2014 and gaining significant momentum since Google and Apple adopted it as part of the Matter smart home standard. Like Zigbee, it is low-power and forms a mesh. Unlike Zigbee, Thread is IP-native: every device on the network has an IPv6 address and communicates directly over standard internet protocols. This removes the dependency on a proprietary coordinator. Thread requires a border router (which can be a small, inexpensive device) rather than a vendor-specific hub. Thread devices are more expensive than Zigbee today, but the gap is closing.
WiFi sensors use your existing wireless network. They are familiar and easy to configure, but they come with trade-offs. WiFi radios are power-hungry, which means WiFi sensors either need mains power or have significantly shorter battery life than Zigbee or Thread equivalents. WiFi networks in care homes are often designed around human-scale devices: laptops, tablets, phones. Adding dozens of sensors can strain access points and create network management complexity. For sensors that genuinely need mains power anyway, such as some radar falls detectors, WiFi is a reasonable choice. For battery-powered environmental sensors, it is rarely the right protocol.
Wired and USB sensors are relevant in specific contexts. A temperature probe in a medication fridge might run wired to a small hub that then communicates over WiFi or Ethernet. Wired connections are reliable and do not consume battery, but installation is more disruptive and costly.
For most care home deployments, a Zigbee-based or Thread-based mesh offers the best balance of battery life, range, installation simplicity, and cost. The critical question is whether the hub or coordinator is vendor-proprietary or whether it supports open protocols.
Falls in care homes cause 40% of all injury-related hospital admissions in older people. The monitoring technology designed to address this has a significant adoption problem, and it is worth understanding why before evaluating alternatives.
Pendant alarms have been the standard approach for decades. A resident wears a button or lanyard device and presses it when they fall. The problems are well documented. Residents with dementia forget they are wearing the device or do not understand its purpose. Residents who fall may lose consciousness or become disoriented and cannot press the button. Many residents refuse to wear them consistently, particularly at night when fall risk is highest. Compliance rates of 40 to 60% are commonly reported, meaning a large proportion of residents are not covered at the moments that matter most.
The more effective approach is non-contact detection, and the technology that has become the category leader is millimetre-wave radar.
Radar-based falls detectors mount on the ceiling or wall and use low-power radio frequency signals to detect human presence and movement without any cameras or wearables. The device tracks body position continuously: whether a person is standing, sitting, lying in bed, or lying on the floor. Modern radar sensors achieve sensitivity of 98 to 99% for fall detection, meaning they correctly identify nearly all falls. Crucially, they do not require any action from the resident and do not require the resident to wear anything.
Privacy concerns that apply to camera-based monitoring do not apply to radar. The device does not capture or transmit images. It processes radio reflections locally and outputs only presence and position data. This is a meaningful distinction for resident dignity and for the comfort of families.
Acoustic monitoring, which uses microphones to detect the sounds associated with a fall, can complement radar in specific scenarios. It is less reliable as a standalone system because noise in care environments creates false positives, but combined with radar it provides additional signal confidence.
The limitation of radar is cost. A quality radar falls detector costs significantly more than a pendant system per room. For high-dependency residents or memory care units where falls risk is greatest and pendant compliance is lowest, the investment calculus typically favours radar. For lower-risk residents, pendant or acoustic monitoring may be proportionate.
The key evaluation question for any falls detection system is: what is the false positive rate? A system that generates frequent false alarms causes alert fatigue, and alert fatigue kills the clinical value of any monitoring system. Vendors should be able to provide sensitivity and specificity data from independent evaluations, not just internal testing.
CQC inspections increasingly examine whether providers are using data to support safe care, not just whether care plans exist. Monitoring data is relevant to several specific areas.
Medication safety requires evidence that medicines are stored at appropriate temperatures. CQC's guidance references the requirement to store medicines as per the manufacturer's instructions. Manual temperature logs can evidence this, but continuous electronic monitoring provides stronger evidence and catches excursions that manual spot-checks would miss. An automated temperature log that records every 15 minutes, with automated alerts for excursions, is more defensible than a paper log filled in once daily.
Night checks are a sensitive area. Some residents find frequent night checks disruptive, and there is clinical evidence that disrupted sleep is harmful for residents with dementia. Passive motion monitoring can supplement or reduce the frequency of intrusive night checks while maintaining safety evidence. Any change to night check protocols requires a formal care planning process and family consultation, but monitoring data can support the clinical case for a less intrusive approach.
Infection control requires evidence of environmental management. Humidity monitoring in bathrooms and communal areas supports infection control documentation. Temperature logs in laundry and kitchen areas contribute to hygiene evidence.
Staffing and deployment decisions can be informed by occupancy and activity pattern data. If monitoring shows consistently low activity in certain areas at certain times, that data can inform staffing schedules in a way that is defensible at inspection.
CareGate Analytics brings this monitoring data together with staffing, incident, and care planning records so that the connections between environmental conditions, staffing patterns, and resident outcomes become visible over time, rather than being something you reconstruct retrospectively after an incident.
The monitoring market has a structural problem. Most vendors build proprietary ecosystems: their sensors use protocols only their hub supports, their hub only sends data to their cloud platform, and their cloud platform exports data in formats that are difficult to move elsewhere.
This creates dependency. When you buy a proprietary monitoring system, you are not just buying sensors. You are entering a long-term relationship with that vendor on their terms. If the vendor raises prices, you have limited negotiating power. If the vendor discontinues a product line, you absorb the replacement cost. If you want to add a sensor type the vendor does not offer, you cannot. If a competitor offers a better falls detector, you cannot use it without replacing your infrastructure.
The alternative is a platform approach: a sensor management layer that supports open protocols and can ingest data from sensors made by multiple manufacturers. This is architecturally more complex to build, but it fundamentally changes the economics of monitoring for the operator.
With an open platform, you choose sensors based on performance and price, not on which vendor's ecosystem they belong to. You can replace a sensor category when something better becomes available without touching the rest of the system. Your data lives in a platform you control, not in a vendor's proprietary cloud.
CareGate Sense is built on this principle. It supports Zigbee, Thread, WiFi, and wired sensors from any manufacturer, presents them in a unified dashboard, and connects to the wider CareGate platform for care planning and analytics. If a better radar falls detector comes to market next year, you can add it. If a medication fridge monitor works better for your supplier's specific equipment, you can use it. The platform is the constant. The hardware choices remain yours.
Before committing to any monitoring platform, there are specific questions that will reveal whether a system is genuinely open or whether it is a proprietary ecosystem with open-sounding marketing language.
Which sensor protocols does the hub support? If the answer is a proprietary protocol, or a single standard protocol with no others on the roadmap, that is a lock-in risk. A genuinely open platform supports at least Zigbee and can demonstrate this with third-party sensors.
Where does the data live? If sensor data is processed and stored only in the vendor's cloud, you need clear data processing agreements that comply with UK GDPR, a mechanism to export your own data, and a clear position on what happens to your data if you end the contract.
What is the alert delivery model? Alerts should reach the right people through the right channels: a falls alert should reach the on-duty carer immediately, not sit in a dashboard that nobody checks at 3am. Integration with existing nurse call systems, or a clear escalation workflow, is essential.
How is the system maintained? IoT devices require firmware updates, battery replacements, and occasional hardware faults. Understand what ongoing management looks like and whether that is included in the contract or charged separately. Managed IT support for care homes should include monitoring infrastructure, not treat it as a separate concern.
What does the pilot process look like? Any credible vendor should support a structured pilot before a full building deployment. A pilot lets you evaluate actual detection performance, false positive rates, and staff workflow impact before committing to a contract.
Monitoring infrastructure in a care home is a significant decision. The hardware lives on your walls and ceilings for years. The data it generates informs clinical decisions and CQC evidence. Choosing the wrong platform, particularly one that locks you into proprietary hardware, is an expensive mistake to reverse.
CareGate Sense is designed specifically for care homes that want the benefits of connected monitoring without the lock-in that has defined this market. It supports any sensor, any protocol, and gives you a single place to manage environmental data, occupancy, falls detection, and biomarker tracking alongside the rest of your care operations.
If you are evaluating monitoring systems for your care home or group, we can walk you through the technical options and help you think through the deployment approach before you commit to anything. Start by looking at what CareGate Sense covers, and speak to us if you want to understand how it would work in your specific setting.