Intended for healthcare professionals

Head To Head

Should patients with dementia who wander be electronically tagged? Yes

BMJ 2013; 346 doi: (Published 20 June 2013) Cite this as: BMJ 2013;346:f3603
  1. Rupert McShane, consultant in old age psychiatry
  1. 1Oxford Health NHS Foundation Trust, Oxford, UK
  1. Rupert.McShane{at}

Technology that allows real time tracking is being pushed by authorities to cut the costs of finding patients who have got lost. Rupert McShane says GPS tracking reduces risk of harm for some patients who wander and that ethical risks are slight, but Desmond O’Neill (doi:10.1136/bmj.f3606) thinks that care must be patient centred rather than looking for quick fixes to suit carers

When a previous chief executive of the UK charity the Alzheimer’s Society wrote in a newsletter that if he had dementia he wanted to be allowed to go out and risk getting lost without having a “GPS tag”—a device that uses global positioning systems to determine location—he received a large mailbag from carers telling him, forcefully, that they disagreed.

Anyone who has lost a child in a public place will remember the cold sweat, the fear in the pit of the stomach, and the guilt as you realise that you have failed to keep them safe. It is the same for carers of people with dementia.

If the risk of getting lost is recognised simple precautions can help. A £15 (€18; $23) pair of transmitters will sound an alarm if the person gets separated in a shopping mall. A cheap buzzer can waken a carer when the patient breaks an infrared beam as she reaches for the front door. A MedicAlert bracelet or a card with a carer’s phone number will help. Good neighbours can undoubtedly help, and technology may be useful in coordinating them. For example, Neighbourhood Watch has set up a rapidly growing pilot project using a police community messaging system to marshal volunteers rapidly to a search.1

Benefits of tracking

GPS trackers, whether bespoke or in the form of mobile phones apps such as Find my Friends, are not a panacea, but they do mean that patients can be found more quickly. This is useful for several reasons. Firstly, rapid recovery reduces risk. Typically, carers delay calling for help, wanting to avoid involving the police if possible. Half of all people with dementia who are missing for more than 24 hours die or are seriously injured.2 Fortunately, only 1% of people with dementia die while lost. However, 40% of those with dementia get lost at some point, and about 5% get lost repeatedly.3 It is this 5% who are the most obvious candidates for a tracker. The first episode of getting lost is usually not predicted4 and is often followed by restrictions on freedom and increased observation, reducing the perceived need for a device.3 5 Secondly, when someone with dementia gets lost, it at least doubles the risk of admission to a care home, making it the strongest behavioural risk factor.3 6

Perhaps one reason that wandering has been given so little attention is that, unlike changes in mood or agitation, getting lost cannot be treated with drugs. Anecdotal evidence suggests that an electronic tracker may delay or avert admission.7 8 The evidence that patients are given more freedom once they start using a tracker is contradictory,7 8 but the finding that the device can reduce conflict between patient and carer is plausible and useful.8

The practical and ethical merits of GPS are best illustrated by real examples.7 Take the woman who loves walking in an area where mobile reception is unreliable. Her husband is prepared to take the risk that the GPS tracker might fail rather than lock her in.7 Another example is a man who lives alone supported by visits from professional carers, who make sure he is wearing a tracker each morning. He does not accept that there is a problem, and needs a little cajoling to carry the device, but rapidly forgets that he is carrying it. His daughter gets text alerts from his GPS tracker if he goes outside a defined area. She has sufficient confidence in the system that she will stay at work and pick him up later. Regular calls to the police are no longer needed, and he can be managed at home despite moderately severe dementia.

Of course, GPS is not infallible. When he has finished work each day, a son collects his father from a city street by logging on and finding him. The system works well until one day it fails. It is two days before his father is found in an emergency department on the other side of the city. In retrospect, the son does not think the risk was worth it.

Safety and autonomy

Anyone can buy a GPS device. The practical and ethical risks are not so great that there is great demand for a formal legislative framework. If concerns about abuse and coercion arise, the “safeguarding” framework provides a proportionate way for multidisciplinary teams to investigate.9 10 In balancing safety and autonomy, social work professionals give more priority to autonomy than do families, and tend to avoiding “cajoling” for fear of overstepping the mark into coercion.11

Ascertaining the degree of insight—a key aspect of autonomy—can be difficult. Grown men have been known to insist, “Of course I’m not lost”—and, in the early stages, cognitive impairment may exacerbate this constitutional stubbornness. Privacy is important, but the desire for privacy and isolation should not be confused with getting lost. The risk-benefit balance is different when observation is focused on the vulnerable. This was encapsulated by the patient who likened a tracker to a bell on a Swiss cow but still wanted to carry it.

The “ethical debate” can itself be stigmatising. One carer described being anxious that her neighbours would ostracise her because her husband did not know she was putting a tracker in his coat pocket. “I’m only trying to keep him safe,” she wept.6 It is time to move on—and we should begin by abandoning the term “tagging,” which has stigmatising associations.


Cite this as: BMJ 2013;346:f3603


  • Competing interests: I have read and understood the BMJ Group policy on declaration of interests and declare the following interests: I instigated the Neighbourhood Watch based scheme referred to.

  • Read Desmond O’Neill’s side of the debate at doi:10.1136/bmj.f3606.

  • Patient consent not required (patients anonymised, dead, or hypothetical).

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract

Log in

Log in through your institution


* For online subscription