Best Dementia Wandering Alarms — Door, Bed and GPS Options

"Wandering" is the clinical shorthand, but it describes the outcome rather than the behaviour, and the mismatch matters when you are choosing equipment. People living with dementia who leave the house are usually doing something purposeful. They are going to work at a job they retired from twenty years ago, collecting a child from a school that closed, walking a route that once made sense, or leaving a place that no longer feels like home. Alzheimer's Society uses the phrase "walking about" for exactly this reason, and advises taking the time to understand why the person is walking and what they need.

Two practical consequences follow. First, an alarm that stops the walking without addressing the need tends to escalate distress rather than resolve it — which is why exit alarms work best paired with a change to the environment or the routine, not alone. Second, the destination is often predictable to someone who knows the person's history, which is a far better search asset than any technology.

The scale is not marginal. The Alzheimer's Association states that six in ten people living with dementia will wander at least once, and many do so repeatedly, with risk present at any stage of the disease. Its listed warning signs include returning from a walk or drive later than usual, forgetting familiar routes, talking about past work obligations, wanting to "go home" while at home, restlessness or pacing, and difficulty locating rooms in their own house. The Association also notes that wandering can occur with no warning signs at all, which is the argument for putting something in place before the first incident rather than after it.

Almost everything sold for this problem falls into one of four groups. They are not interchangeable and most households eventually use two.

• Exit alarms — a sensor on the front or back door, or a mat by the threshold, that chimes locally or notifies a phone when the door opens. Catches the leaving.
• Bed and chair sensors — a pressure pad that alerts when the person gets up and does not return within a set time. Catches the getting up, usually at night, before the leaving.
• Floor mats and beam sensors — placed at a bedroom door or the top of the stairs, alerting on movement through a specific point. Sits between the two above.
• Worn locators — a GPS watch, pendant, insole or keyring that reports position, usually with a geofence that alerts when the person leaves a defined area. Catches where they went.

The important structural difference is who gets the alert. A local chime helps only if someone is in the house. A notification to a phone helps only if that phone is looked at. A monitored device routes to a call centre that will act. Ask which of the three you are buying, because vendor pages routinely use the word "alert" for all of them.

An exit alarm is the cheapest useful thing in this category and, for most households, the first thing to fit. The mechanics are simple: a magnetic contact sensor on the door frame, a chime unit, and optionally a smart-home hub that turns the chime into a phone notification.

Where they work well. Night-time exits in a household where a carer or spouse sleeps in the house. A chime in the hallway wakes the carer within seconds of the door opening, which is a far shorter interval than discovering an empty bed at 6am.

Where they fall down. A person living alone gains nothing from a local chime — the alert has to travel. Alarm fatigue is real: a sensor on a door that opens twenty times a day for legitimate reasons stops being read within a fortnight. And a determined person will use a window, a back gate, or the door the sensor is not on, so a single sensor is a partial measure.

The thing to avoid. Locking the person in. Beyond the fire risk, restricting a person's freedom of movement engages deprivation-of-liberty rules in the UK and equivalent protections elsewhere. Alzheimer's Society advises that where a person lacks capacity, social services should be consulted before doors are locked. A door alarm is not the same as a locked door, legally or ethically, and that distinction is the reason the category exists.

These are the most underrated devices on the page. A pressure sensor under the mattress or on the chair detects the person getting up and, after a configurable delay, raises an alert if they have not returned. That delay is the whole trick: it lets an ordinary bathroom trip pass silently while flagging the one where the person did not come back.

They address two problems at once — night-time wandering before it reaches the door, and unwitnessed falls on the way to the bathroom, which is one of the most common serious incidents in older households.

Practical notes. Placement matters more than brand: a mat that is easy to step around does nothing. Battery-powered units need a documented replacement routine, because a silent dead sensor is worse than no sensor because it manufactures false confidence. And in a household where the person shares a bed, mattress sensors need setting up carefully to avoid constant triggers.

Many UK councils supply and monitor exactly this equipment through telecare schemes; it is worth an assessment before buying privately. See free personal alarms for the elderly for the routes.

A worn locator answers the question the other three cannot: where did they go. The formats are a wristwatch, a pendant, a keyring or fob, and insoles or clips that hide in clothing. Most support geofencing — an alert when the person crosses a defined boundary — and most offer a location history.

Four things decide whether one of these works in practice, and none of them is accuracy.

Will it be worn? A device that is not on the body when the person leaves is decoration. Watches and jewellery-format devices survive far better than anything that has to be remembered as a separate object. Insoles and clothing clips solve the wearing problem and create a different one, discussed in the consent section below.

Battery life. Continuous GPS is power-hungry. A tracker that reports every minute may last a day; one that reports every fifteen minutes may last a week. The right setting is the one that survives a household's actual charging habits, not the one with the best specification.

Who receives the alert, and what do they do? A geofence alert to a phone at 2am is only useful if the phone is not on silent and the recipient can act. Some services route to a monitoring centre instead; that costs more and is worth it for households with no nearby responder.

Coverage. These devices need cellular signal. Rural areas, basements and some buildings will produce gaps, and a gap is indistinguishable from a stationary device.

Alzheimer's Society is direct about the precondition: a person needs to consent to tracking before it is switched on. Our comparison of GPS trackers versus a daily check-in covers where tracking earns its intrusion.

Every carer support organisation puts these ahead of technology, and they are right, because they work when the battery is flat.

• Identification on the person. A card in the wallet, an engraved bracelet, or a Helpcard of the kind Alzheimer's Society produces, giving a name and a phone number. This is the single highest-value item on this page and it costs almost nothing. Most people who walk about are found by a member of the public, not by a search.
• Tell the neighbours and the local shops. With the person's agreement where possible. A shopkeeper who knows to phone you is faster than any geofence.
• Write down the likely destinations. Former homes, former workplaces, a church, a particular bus stop. Give the list to whoever will search. The Alzheimer's Association's observation that many people are found within 1.5 miles is what makes a short, specific list so effective.
• Address the trigger. Walking about often follows boredom, pain, needing the toilet, or a time of day associated with an old routine. A structured afternoon activity resolves more incidents than a sensor detects.
• Keep a current photograph. Recent, clear, full-face. Police will ask for one immediately and searching a phone under stress is not the moment to find out you only have holiday snaps from 2019.

Our own page on night-time wandering goes further into the overnight pattern specifically.

Fitting a tracker to an adult is a decision about somebody else's liberty, and it deserves more than a checkout page.

In England and Wales, the Mental Capacity Act 2005 sets the framework. Its principles are that a person must be assumed to have capacity unless it is established that they lack it; that a person is not to be treated as unable to make a decision merely because it is an unwise one; and that where capacity is genuinely absent for a particular decision, anything done on the person's behalf must be in their best interests and must be the least restrictive option available (Mental Capacity Act 2005, section 1). Capacity is decision-specific and can fluctuate — a person may lack capacity about finances and retain it about whether they will wear a watch.

Alzheimer's Society's position follows the same logic: the person would need to consent to tracking before it is switched on; if they have capacity they can decide to do things even if that puts them at risk; and where they do not, all decisions must be made in their best interests, with the person involved in the discussion as far as possible.

In the United States there is no single national standard. Capacity, guardianship and consent are matters of state law, and the practical mechanism is usually a healthcare proxy or durable power of attorney executed while the person can still do so, plus advice from an elder law attorney licensed in the relevant state. Assume variation rather than a uniform rule.

The concealment question deserves its own answer. Insoles and clothing clips exist precisely because they can be used without the person knowing. Sometimes a covert device is genuinely the least restrictive option available for someone who lacks capacity and would otherwise be locked in or moved — that is a real and defensible best-interests judgement, and it should be documented as one, ideally with a social worker or clinician involved. What it should not be is a private decision made at 11pm because the alternative conversation is difficult. If you cannot write down why the covert option is less restrictive than the alternatives, that is the answer. Our page on the ethics of informed consent in monitoring goes further into this.

A short list that catches most of the regret in this category.

• Who receives the alert, on what device, and what happens if they miss it? Local chime, phone push, or staffed centre — three very different products.
• Is there a monthly fee, and what does cancelling involve? Notice periods and leased equipment are the usual traps. Fall detection, where offered, is frequently a separate recurring charge.
• What is the real battery life at the settings you will actually use? Not the best-case figure.
• Will it survive water, a shower, and being slept in?
• Does the person accept it? The most sophisticated device with a 5% wear rate is worse than a laminated card with a phone number on it.
• What is the plan for the first ten minutes after an alert? Write it down, give it to everyone, and rehearse it once. This matters more than the device.

On price: figures in this market move and many vendors do not publish them, so check the vendor's own page on the day you buy rather than trusting any comparison article, including this one. As a UK reference point that is published openly, Age UK Trading listed its Personal Alarm Watch at £99.99 setup plus £24.49 a month and its Taking Care GO GPS alarm at £79.99 setup plus £25.49 a month, ex VAT, when we checked in August 2026; Age UK states that 94% of its personal alarm customers do not pay VAT because they qualify for disability VAT relief.

We make a daily check-in app, and it is not a wandering solution. We would rather say that here than have you find out after paying.

I'm Alive is a daily check-in app. You confirm you're okay once a day; if you don't, the people you chose are told, with your location. The daily check-in is free (Try It, $0) and works on iPhone and Android anywhere in the world. Automatic alerts to your contacts start on Protect Me ($29.99/year), with location on the move on Protect Me On The Move ($39.99/year); there is a one-time Stay Connected lifetime upgrade at $4.99. Check-first phone fall detection is live on iPhone only (app v2.3.1) on Protect Me and above, stays off until you switch it on, and only runs while the app is open — it is not a background service. It alerts your own trusted contacts; it never calls emergency services, and it is best-effort, not a medical device. There is no Android fall detection and no Apple Watch fall detection. We do not operate a 24/7 monitoring centre. The app interface is currently in English only.

A check-in asks a question once a day and alerts your contacts if it is not answered. Against wandering, that is the wrong timescale by an order of magnitude — someone who leaves at 11pm needs to be missed within minutes, not by tomorrow's prompt. And a check-in depends on the person reliably responding, which is exactly the capability that mid-stage dementia erodes. If wandering is your primary concern, buy a door alarm, a bed sensor, or a worn locator, and put an identification card in the person's coat today.

There is one place we do fit. Carers in this situation are frequently older, frequently alone in the house overnight, and monitored by nobody. If you are the person who would notice, it is worth asking who would notice you. Related reading: dementia monitoring systems by stage and the full monitoring taxonomy.