Abstract
Summary
There are about 5 million people in Europe who have dementia, approximately half of whom need daily care. A common reason why dementia sufferers are admitted to long-term care is because of “wandering”, i.e. leaving home without informing a carer, thereby potentially putting themselves at risk. Common methods of managing wandering include locking doors or alerting carers when a door is opened. A new method of managing wandering is by using electronic location devices. These depend on the satellite-based global positioning system (GPS). People can wear a location device in the form of a watch or pendant, or carry it like a mobile phone. This offers affected individuals the possibility of safe walking, with the reassurance that they can be found quickly if lost. However, it is not known how effective this method is and its use raises questions about safety and individual civil liberties. GPS location is a potentially useful method of managing wandering in dementia and there is considerable pressure on caregivers from commercial organisations to adopt the technique. Research is therefore required to determine which people are best suited for such devices, how effective they are in practice and what effect they have on important outcomes.
Introduction
Dementia is common and increasing, and there are now approximately 750,000 people in the UK, five million in Europe and about 24 million worldwide with diagnosed disease.1 However, as many as three quarters of all cases may be undiagnosed.2 As the population ages, the incidence and prevalence are expected to increase rapidly, so that by 2021 there will be approximately one million affected people in the UK.3 Currently, approximately 64% of the people in the UK with dementia need care daily at a home, many of whom require constant care or supervision. Approximately 36% are in long-term care.3 Most people with dementia wish to remain at home, and health and social care efforts try and achieve this, particularly as long-term care is increasingly scarce and expensive.4,5 A key reason why dementia sufferers are admitted to long-term care is because of “wandering”, i.e. leaving home without informing a carer, thereby increasing the risk of accidental injury.6 Up to 40% of people with dementia wander, with just under 10% wandering regularly.7 People may be seriously injured while wandering; carers are made anxious and may spend many hours searching for them.6 In addition the police often become involved in searching. Often, however, the person stays close to home, follows familiar routes, is at relatively low risk of harm and will be able to find their way home.
While non-cognitive effects of dementia such as wandering are a major cause of carer stress,7 this must be balanced against the potential benefits in terms of physical exercise and the perception of autonomy afforded by so called “safe walking”.8,9 Indeed, for people with mild-to-moderate cognitive impairment, short walks in a familiar area may be a good source of social contact and allow informal supervision by neighbours and local shopkeepers. In addition it is a good source of exercise. Problems may however arise if the normal routine is upset through a physical change in the environment necessitating a deviation from the usual route (for example, road works) or an illness (such as a urinary tract infection) leading to further cognitive impairment. A distressing episode may lead to a loss of confidence and this may be compounded by familial pressure to stay at home.
Non-GPS based methods of managing wandering
The most appropriate approach to managing wandering and promoting safe walking depends on, amongst other things, the degree of cognitive impairment, familiarity of the person with his or her surroundings, history of previous episodes of getting lost, and the availability of family or friends to provide help if necessary. For the least impaired people, a simple mobile phone with pre-programmed numbers to call if they get lost may be sufficient. This is often enough to reassure both them and their carers. Clearly, they still have to remember to take it with them and if they get lost be able to use it. However, it is possible that even with a mobile phone they may not be able to describe where they are. Approaching strangers for help may further compound any risks. Such individuals with relatively mild disease may thus potentially have most to gain from the global positioning system (GPS) location technologies described below.
For the most impaired people, no degree of unaccompanied walking outside the house is safe, either because they have poor road sense and may be injured or because they very quickly get lost and relatives have to spend much time looking for them. The only solution is to provide barriers to leaving the home, such as by locking the door, making it complicated to open (e.g. requiring a code to be entered on a number pad) or providing floor patterns which discourage crossing a threshold.10 These have the disadvantage of impeding escape in emergency situations such as a fire.
Electronic devices on doors can detect when they have been opened and alert the carer. Such devices may also provide a voice prompt, i.e. a message previously recorded by somebody the client is familiar with. The devices can be programmed to be time sensitive, for example informing the person that that “it is the middle of the night” and advising them to stay at home. Furthermore, they can be directly linked to a relative or through a call monitoring system to social care teams or a commercial provider. If a call-monitoring centre is involved, staff can either try to talk to the person through the community alarm unit to persuade them to come back indoors or ask a nearby responder to attend. However, devices on doors cannot record whether the person has actually left the house, or is very close by, for example, in the garden. Another drawback of such systems is that occasionally they reveal a previously unrecognised high frequency of night-time excursions, many of which will be minor, but which are nonetheless sufficient to heighten relatives’ anxiety and which may as a result trigger admission to long-term care.
Evidence for non-GPS methods
A recent review explored non-pharmacological interventions for the prevention or management of wandering in the domestic setting.11 However, the review found no randomised controlled trials or other high quality research that demonstrated the effectiveness of interventions focusing on environmental change, including music therapy, bright light therapy or psychological or exercise approaches such as reality orientation, physical therapy, occupational therapy and therapeutic touch. A Cochrane review exploring physical non-pharmacological interventions in the domestic setting concluded that there was an urgent need for high quality trials investigating non-pharmacological methods of managing behavioural problems such as wandering.12
Potential for GPS location
Against this background one solution to wandering or encouraging safer walking is the use of electronic location. This is commonly based on GPS. The GPS is a global navigation system based on a constellation of orbiting satellites. It provides location information anywhere on earth where there is an unobstructed line of sight to four or more GPS satellites. A GPS receiver measures the distance to each satellite. These distances, along with the satellites’ locations, are used to compute the position of the receiver. In a location device, this position can then be sent by mobile phone signal to the location service. The exact coordinates of the receiver (accurate to a few metres) can then be superimposed on an electronic map.
People can wear a location device in the form of a watch or pendant, or carry it like a mobile phone. There have been considerable advances in the miniaturisation of such devices in recent years. In its simplest form, a person with mild cognitive impairment can carry a device which they switch on if they get lost. This initiates a call for help, accurately pinpointing their location to within a few metres as long as they are outside and visible to satellites. However, if the person is indoors or out of view of the satellites, then GPS location will not work. In these circumstances, some systems rely on triangulating mobile phone signals. This provides a much less accurate position. Modern smart phones can use a mixture of GPS and triangulation to track a person's whereabouts; mobile phone apps aimed specifically at locating people who have become lost are becoming available.13
Occasionally, GPS location systems are used as an assessment tool to reassure carers about usual walking behaviours, but more commonly they are sold for ongoing monitoring. Searches can be triggered by the carer, for example if the client is late home from shopping. Carers can either contact an alarm call centre or log-on themselves to the computer system. By sending a mobile phone signal to the device it can switch on satellite location and the person can be pinpointed on an electronic map.
A more actively monitored solution involves the use of “geofences”. Working together with the person and their carer(s), a detailed plan is made of usual routes and when these individuals usually take them. This can be restricted to the person's own garden, for example, or tailored to a familiar bus route to a park. If the client leaves the area surrounded by the geofence, or sets out at an inappropriate time, carers are alerted, the GPS receiver is switched on and their location can be determined.
Location systems require someone to be available both to monitor and to act on alerts, which may include phoning the individual to check that they are alright, or finding and bringing home the wandering person. The technology can be managed entirely by a person's family or by a system involving a call-centre who either alert the family or other carers or can send people to search. Occasionally, the police are involved in cases where cognitively impaired people go missing. Accurate location systems can be linked to smart phones which can have a picture of the person transmitted to them along with personal data such as the diminutive of the name by which he or she likes to be called or information about their background such as their daughter or son's name which can be used to reassure them.
Assessment
Many carers (and occasionally people with cognitive impairment) approach social services or commercial companies to see if a device might help them. Guides have been developed to help people decide if the technology is suitable, and if so what type of device or service would be best for them.14 Assessment includes a full history of concerns about the client (previous times when they have become lost, wandering episodes, instances of injury) and an assessment of risk to the individual. It is important to establish normal routines for the client so that use of the device does not end up curtailing useful exercise and social interaction. It is essential to determine what support (either family or social services) exists to monitor and search for clients should they become lost or wander. It is also important to establish mental capacity as part of the assessment. If clients do not have the capacity to decide then it must be clear that the proposed solution is the least restrictive and in their best interests. Providing a GPS locating device might be seen as trespassing on the person's rights if applied when the person is not at high risk of becoming lost. Some social services provide devices for a limited period for carers to see if they like them and find them useful. In any case it is important for a review to take place after a few months of use.
GPS technologies and the evidence
GPS devices are already being used by people with dementia and some manufacturers make inflated claims about their effectiveness and safety with respect to enabling people to continue to live at home. Although small pilot studies have shown that the technology can help to locate people quickly15 no high quality trials have been conducted to explore the safety, or the effects on people with dementia, their carers or resource use. In particular, little attention has been paid to the potentially serious risks of relying on such devices. For example, while the prospect of maintaining or even increasing safe walking is attractive, it is not known if this increased activity will come at the expense of increased accidents, as this group of people may have reduced awareness of the risks of traffic.
While such devices seem relatively popular with many family carers,16 professionals, in particular, have some reservations about the impact on civil liberties.17 Civil liberties do not seem to bother family carers as much as professionals, and this difference is more marked with older carers.18 However, the potential for unnecessary invasion of privacy and bullying remains.
Others have suggested that the number of people likely to benefit and take up such a service is small (around 25% of wandering people), partly because for some people no degree of unsupervised walking outside is safe, and that the use of such devices may lead to an increased risk of accidents.15 Patient groups recommend that the devices should only be used for people who choose to leave the house with a purpose and that they should not be used for maintaining people within a secure environment.9
A qualitative study of a commercial product carried out by White et al.19 revealed that families often used the devices as a back-up rather than the main method of managing people who became lost or wandered. This was largely because of the unreliability of the system. However, despite this, both users of the system and their carers felt reassured by the device and in general felt that independence was enhanced by its use.
McShane et al. reported a small (n = 24) feasibility trial of an early location device (a simple radio transmitter) which was poorly tolerated and continuously used by only one-third of people.18 However, the device was large and primitive compared with those now available. Miskelly carried out a small (n = 11) technical trial of a GPS based system.15 However, by modern standards this too was somewhat bulky (weighing about 500 g) which was one of the main reasons given for clients discontinuing its use.
Although more portable, the smaller more modern devices have limited battery life and therefore have to be charged regularly. The client can do this, if they have the mental capacity, or a family member who lives with them, or a carer, although this necessitates a daily visit. Limited battery life affects the acceptability and the cost-effectiveness of the intervention. The battery life depends on the frequency with which the device uploads location data. Clearly the more frequently this information is transmitted the easier it is to find someone. Upload frequency can change if the person moves outside a geofence, but it may be some time before this is detected if transmissions are infrequent. System alerts can be set up if the GPS signal is lost or the battery is failing. Speed detection algorithms can be employed to detect if the client gets on to a bus or into a car.
The lighter devices can be integrated into a mobile phone, or worn as a removable watch, key-ring or pendant for people who can be relied upon to remember to put it on. For the more cognitively impaired, there is a watch-based device with a secure fixing. These watch-based devices are bulkier than normal watches, looking rather like the ‘action’ watches that sportsmen or young people might wear, and some potential clients may therefore not like them, Figure 1.
GPS tracker watch (image courtesy of Everon Oy/Ab, Finland)
Ethical and legal challenges
In the UK and other countries health and social service organisations are beginning to set up services that offer a variety of models of care. These range from helping carers and clients to decide if they are suitable for technology and which particular types of technology are most appropriate for the problems they have, to a complete service which involves carers and police who can be mobilised to find wandering clients and return them to their home. The use of location systems may be perceived as similar to the electronic tagging used for imposing home curfew on felons (although most use a different technology) and can therefore be emotive. The use of language is important and the word “locator” is preferable to “tracker”. While some people view these systems as reducing autonomy, for others the converse is true as their autonomy may be increased.8 Nevertheless, it is important that the person who is to use the locator has his or her mental capacity to consent assessed, and that their best interests are taken into account as well as those of their carers. It is also important to take into account appropriate limits of power of attorney when decisions are being made. Last, there may be concerns about who has access to the tracking information. It is not always appropriate to know where someone is at any time and access to the client's position should only be provided when there is a concern about their whereabouts and should be restricted to trusted individuals. A log should be kept of all accesses to the location website.
Future research
Given the considerable pressure that caregivers are coming under from commercial organisations to adopt these systems, there is an urgent need to determine for which people such devices are likely to be acceptable and most effective, and what effect the use of such devices may have on important outcomes such as the time taken by carers and/or the police to find lost individuals, carer stress and anxiety, accidents, delay of admission to long-term care and the effect on service resource use. This research will be challenging, particularly with regard to ethical issues such as consent and capacity20 and in choosing appropriate instruments for measuring outcomes. Qualitative research into attitudes and the experience of using more modern devices has been completed19 and is also under way in Scotland.21 Recruitment, particularly to randomised trials, may be difficult in an environment where there is sustained pressure on local authorities and carers from commercial companies to roll out this technology, albeit with limited evidence of efficacy. Even if randomised trials are not possible there is the potential to use an interrupted time series or a step-wedged trial design.22
Conclusions
GPS location is a potentially useful method of managing wandering in dementia. However it is important that the people to use it are selected with care. The overall effect of the technology in terms of safety, reduction in stress for carers and people with dementia and on use of resources has yet to be established.
Footnotes
Acknowledgements
We thank Heather Milne, Lucy McCloughan, Donna Fleming and Alison Anderson for their helpful advice. We also thank the Chief Scientist Office of the Scottish Government for support. B McK is supported by NHS Lothian through the Edinburgh Health Services Research Unit.
