Abstract
This paper details a review of the literature on the use of electronic aids for prospective memory for people with dementia. Key findings of the review are that: electronic memory aids show potential for supporting people’s prospective memory but the devices and software applications need further development in order to function reliably; sample sizes of studies are often very small, limiting the generalisability of their findings; few studies of devices are conducted in users’ home environments; and most of the studies focus on the effectiveness of the electronic memory aid, rather than outcomes for users, such as improved daily functioning, quality of life, or social connectedness. The review concludes that future studies with robust devices are required that explicitly focus on the varying needs and capacities of people with dementia, in order to generate additional evidence for the effectiveness of electronic memory aids for this cohort.
Introduction
In recent years, published academic literature on electronic assistive technology for older adults has burgeoned, arising partly from technological developments in personal computing and expanded access to high-speed internet, but also fostered by policy interest amongst governments aiming to assist older adults, including those with dementia, to live at home for as long as possible.
For many people with dementia, being in their own home environment is highly valued and important to their quality of life, with the home providing a sense of familiarity and security (Evans, Price, & Meyer, 2016; Svanström & Sundler, 2015). Additionally, people with dementia are reported to spend a significant proportion of their time at home (Evans et al., 2016). However, they often face significant challenges completing activities of daily living in their home environments (Dröes, van der Roest, van Mierlo, & Meiland, 2011; Duane, Brasher, & Koch, 2013; Evans et al., 2016; Harris, 2006; Khosravi, Rezvani, & Wiewiora, 2016; Parikh, Troyer, Maione, & Murphy, 2015; Svanström & Sundler, 2015). Qualitative studies have shown that people with dementia are open to using electronic assistive technology to assist them in their everyday lives (Nygård, 2008; Rosenberg & Nygård, 2014). However, little is known about the ways in which this technology may support them to remain living at home (Boger, Quraishi, Turcotte, & Dunal, 2014).
Assistive technology devices for people with memory deficits are sometimes referred to as “assistive devices for cognition”, which are defined as “any technology which compensates for cognitive deficit during task performance” (Gillespie, Best, & O'Neill, 2012, p. 2). These devices can address four areas of cognition: (a) prospective memory such as remembering to attend upcoming appointments; (b) calculation; (c) memory of past events; and (d) sequencing of complex behaviours (Gillespie et al., 2012). This review focuses on the first of these – prospective memory aids, and specifically those that are electronic, such as smart phones, personal computers, and digital voice recorders (Dewar, Kapur, & Kopelman, 2016).
Prospective memory aids are thought to be useful for assisting people with dementia with time orientation, keeping appointments, attending social activities, and generating a sense of control over daily activities (Arntzen, Holthe, & Jentoft, 2016). As well, these aids may reduce interpersonal conflict between carers and people with dementia, by lessening demands on carers to provide regular reminders (Arntzen et al., 2016).
Methods
This paper reports on a focused literature review, conducted in July 2017, into the development and use of electronic memory aids for prospective memory for people with dementia. The aim of the review was to gain a sense of the size, scope, and quality of published academic literature and to provide recommendations for further research (Grant & Booth, 2009).
This review represents an original contribution to the field in two ways. Firstly, the review focuses exclusively on prospective memory, rather than addressing all types of memory; and secondly, it focuses on studies conducted with people with dementia, rather than studies of memory aids for people with acquired brain injury, or studies that combine acquired brain injury and dementia into one study population. There is a clear need for studies into electronic memory aids that are specifically designed for people with degenerative diseases including dementia, due to decline in their capabilities over time, separate to people with acquired brain injuries (Jamieson et al., 2014). This review is intended to inform the future design of studies for the development of “context-aware” electronic memory aids – that is, technological devices which reflect the particular needs, living situations, capabilities, and values of people with dementia (Mihailidis & Fernie, 2002, p. 173).
Papers on empirical research published between 2000 and 2017 were located via Google Scholar and Web of Science, using the Boolean search terms: (1) “electronic memory aids” AND dementia, (2) “reminder aids” AND dementia, and (3) “prospective memory aids” AND dementia. These searches generated 308 results in Google Scholar and three results in Web of Science. The search in Web of Science was subsequently widened by using the search term “memory aids and dementia”, which produced 29 results. Duplicate papers were removed, and each paper was reviewed by title and abstract to determine if it reported findings from an empirical study on the development or use of an electronic prospective memory aid for people with dementia. Literature reviews and overview articles were excluded, and the following exclusion criteria were applied to the focused review: (1) the paper was in a language other than English; (2) all of the study’s participants had conditions other than dementia; (3) the study was conducted with healthy participants; (4) the study did not address prospective memory; and (5) the study was not concerned with an electronic device.
Results
The focused search process identified that since 2000, there have been only 16 published empirical studies of prospective memory aids for people with dementia. These studies are summarised in Table 1. Generally, these studies provide evidence that people with dementia are capable of learning to use electronic memory aids, such as touch screens (Davies et al., 2009; Leng et al., 2014), and these devices are potentially useful to people with dementia (Boman et al., 2016; Topo et al., 2007).
Empirical studies of electronic prospective memory aids.
However, four key areas for improvement are noted. Firstly, the majority of the studies were conducted at an early stage in the development of devices which adversely affected their functioning and reliability, thereby confounding the results of the studies. Secondly, researchers have primarily focused on assessing the effectiveness of the electronic memory aids, rather than evaluating user outcomes such as improved daily functioning, quality of life, or social connectedness. Thirdly, very few of these devices have been trialled in the home environments of people with dementia. And finally, the studies' sample sizes are often very small, limiting the generalisability of their findings. Each of these areas for improvement are discussed below and inform recommendations for future research.
Potential usefulness of devices
A large series of studies conducted on aids for people with dementia under the auspices of the ENABLE Project has informed further development of aids for people with dementia, by designing user-led methods for ascertaining the needs, values, and expectations of potential users. Six novel aids were designed and developed, including one prospective memory aid, in collaboration with people with dementia and their carers (Hagen et al., 2004; Orpwood et al., 2004, 2007, 2010; Sixsmith, Gibson, Orpwood, & Torrington, 2007).
A study from the ENABLE Project found that an earlier-developed prospective memory aid – the Night and Day Calendar – was useful for people with dementia (Topo et al., 2007). In this study, 50 people with dementia and 50 carers assessed this device, consisting of a digital display of the day and date, and whether it was morning, afternoon, evening, or night. After using it for three months, 78 percent of people with dementia reported that they found the device useful and carers reported that it was useful for 73 percent of people with dementia (Topo et al., 2007).
The Smart Assisted Living Involving CareGivers++ (SALIG) project developed an electronic memory aid for people with either acquired brain injury or dementia, including a calendar and voice reminder system for medication and appointments (Boman et al., 2016). Potential users of the device were included in the design process, and an early mock-up of the tablet was evaluated, during which people with cognitive impairment, health professionals, and carers were shown the device’s functions and then interviewed on their views (Boman et al., 2016). Overall, participants were satisfied with the multi-functionality of the device, its ability to be configured and updated remotely by professionals and carers, and its form as a tablet (Boman et al., 2016). The findings of this study demonstrate the potential usefulness of electronic memory aids for people with dementia. However, because the study was conducted at an early stage in the development of the device, the researchers were unable to assess its impact on remembering medication and appointments.
Technical difficulties with devices
Several studies have reported technical difficulties in their findings which have made it difficult to assess the acceptability and outcomes of those devices. A study conducted in the Netherlands, Germany, and Belgium trialled the newly-developed Rosetta system, which is a combination of three previously-developed assistive technology systems (Hattink et al., 2016). It provides people with dementia with daily appointment reminders via the Elderly Day Navigator, and a surveillance system via a video touch screen or mobile device. The system can be remotely configured and updated by formal and informal carers. In this study, 42 people with mild cognitive impairment and 32 informal carers felt the Rosetta system had potential but were critical of the device itself, which was still under development. Criticism related to its usability and reliability, with nearly all participants experiencing unstable or malfunctioning software systems (Hattink et al., 2016).
A further study also reported technical difficulties; it entailed the development of a wireless touch screen memory board in a residential setting, which was designed to support resident’s memory of their daily activities (Kerkhof et al., 2015). Interviews and focus groups with residents, carers, and staff found that the device effectively supported the memory of people with dementia. However, the usability and reliability of the device were criticised in relation to software installation errors, difficulty of use, and a lack of knowledge of users (Kerkhof et al., 2015). As well, the study’s findings did not allow an assessment of its applicability to those with dementia living in their own homes.
Similar challenges with the functioning of newly-developed devices were experienced in separate studies of three other devices – the COGKNOW Day Navigator, MemAid, and Memojog. The COGKNOW Day Navigator consists of a touch screen, a mobile device, sensors, a home hub, and a central server (Mulvenna et al., 2010). A user-centred development study, consisting of three cycles of one year each, was conducted with 15 people living with mild dementia, in order to align the COGKNOW’s functions with users’ needs. The devices were installed in people’s homes. The first two cycles of the research assessed its useability and usefulness in supporting memory and social engagement, and the third cycle focused on its impact on daily life. Participants frequently experienced problems with the technology, and in the third cycle of the research, no significant impact on their quality of life was found. The researchers recommended that further impact studies be conducted once the system is more robust and stable (Mulvenna et al., 2010).
Developed in the United States, MemAid is an Android application for use with mobile phones, which includes a reminder system as well as an alert for “wandering” behaviour and a memory game (Chen et al., 2015). The researchers evaluated MemAid for functionality and robustness, and identified a series of improvements to be made to the application. No information was reported on the effectiveness of the reminder system; instead, the findings focused on the device’s spoken word recognition capacity, memory game, and GPS tracking system (Chen et al., 2015).
Dundee University in New Zealand developed the Memojog memory aid, an interactive communication system which delivers text-based prompts to users (Morrison et al., 2004). Carers and staff can remotely input reminders to the device via mobile phone technology. The system was evaluated with six participants with severe progressive degenerative memory problems over a 12-week period. Participants liked the usability and multi-functionality of the device but experienced problems with the reliability of the mobile phone coverage in the test areas, and the sensitivity of the touch screen (Morrison et al., 2004). These technical difficulties may be less relevant to newer studies, as both mobile phone coverage and touch screen technology have improved since this research was conducted. However, its effectiveness depends on whether the patient remembers to check the device.
Controlled environments of studies and heterogeneous needs of participants
Several of the studies have been conducted in a controlled environment under test conditions, rather than with “naturally occurring” tasks in users’ homes. A study with five people with dementia, which employed an electronic memory aid to record and play back vocal reminders of appointments or tasks, found it was effective in improving memory functioning of people with dementia, in comparison to either using written reminders or no reminders at all (Oriani et al., 2003). However, its generalisability has not yet been assessed.
Expanding on this, another study conducted under controlled conditions developed a new reminder system for people with dementia which primarily used a Virtual Agent user interface – i.e. a “chat bot” or simulated person who conversed with the person with dementia – as well as a touch screen graphical user interface (Tokunaga et al., 2016). The device was trialled with 17 hospital patients with dementia and their carers, and the study found that participants could use it and interact with the Virtual Agent, at least to some degree (Tokunaga et al., 2016). However, this study did not assess the usefulness of the device for prospective memory, evaluate outcomes for users, or trial its usefulness in home environments.
An effectiveness trial of different computer-based memory aids with 16 older people living in supported accommodation found that, although memory aids proved efficacious in controlled conditions, it was challenging to introduce them for daily use with older adults, who found the devices too complicated and required assistance to use them (Collerton et al., 2014). Participants also thought the memory aids were sometimes a poor fit for their individual needs, with some preferring a more complex and sophisticated device than others, depending on their level of cognitive function (Collerton et al., 2014). The findings of this study point to the importance of designing devices which are adaptable to the heterogeneous needs of people with dementia, as well as to their declining cognitive function over time.
Small sample sizes
Several of the published studies have employed very small sample sizes, which limit the generalisability of their findings. A Japanese team of researchers evaluated the use of a voice recorder manufactured for the general population which can deliver spoken reminders at specified times and days (Yasuda et al., 2002). In contrast to the study by Dewar et al. (2016), the main advantage of the device was its usability; participants in the study did not require training in its use, as it was pre-programmed and worked spontaneously. The study evaluated its use with eight people with prospective memory impairments, who were requested to engage in diary writing, letter writing, or physical exercise via a verbal message automatically delivered by the device. The effectiveness of the device was limited: five of the eight patients completed the tasks as requested by the device, whilst the other three did not, which the researchers speculatively attribute to the type or stage of their cognitive impairment (Yasuda et al., 2002).
A very small longitudinal study of a smartphone application called AP@LZ was conducted over 24 months with only three people with dementia. The application was designed to help people with dementia compensate for their memory problems, including remembering appointments. The study found that participants were able to learn the skills necessary to operate the application, and for one of the three cases, caregiver burden was reduced (Imbeault et al., 2016). However, with such a small sample size, it is difficult to assess the effectiveness of the device and to generalise these findings to the wider population.
Discussion and recommendations for further studies
The conduct of this review was constrained by the diverse nature of the published studies, in terms of their aims, design, sample sizes, and methods. The heterogeneity of these studies meant that the quantitative data could not be compared or combined to conduct a meta-analysis of the findings; this is a significant limitation, which is also noted in a recent systematic literature review of assistive technology for social participation in older adults with dementia (Pinto-Bruno, García-Casal, Csipke, Jenaro-Río, & Franco-Martín, 2016).
Despite this limitation, the studies collectively provide promising evidence for the feasibility of people with dementia using prospective memory aids. Although this cohort generally has greater difficulty using technology than those without dementia (Malinowsky, Almkvist, Kottorp, & Nygård, 2010; Nygård & Starkhammar, 2007; Rosenberg, Kottorp, Winblad, & Nygård, 2009), two modifications to technological interventions may effectively address these difficulties. A recent controlled clinical trial found that participants with a progressive memory disorder, including dementia, may require more intensive training conducted across more sessions, in order to use memory aids effectively over time (Dewar et al., 2016). These findings are consistent with those elsewhere which identify the value of training for users (Mason, Craig, O’Neill, Donnelly, & Nugent, 2012). To further assist this cohort, regular reminders about how and when to use the technology may be required to maximise its use (Cahill, Begley, Faulkner, & Hagen, 2007; Oriani et al., 2003).
Significantly, this review reveals that the evidence base for electronic prospective memory aids is still under development. Studies of the development of new electronic memory aids have experienced difficulties with the usability and reliability of the devices, which has clouded assessments on their effectiveness and longer-term outcomes for users. Accordingly, much of the research into electronic memory aids is focused on the design process, users’ adoption of the device, its usability, or its efficacy in laboratory environments rather than in the homes of people with dementia (Chaurasia et al., 2016; Greenhalgh et al., 2013; Scherer, Sax, Vanbiervliet, Cushman, & Scherer, 2005; Zwijsen, Niemeijer, & Hertogh, 2011). As well, the small sample sizes of many of the studies limits the generalisability of the findings.
These review findings are consistent with those of an earlier, systematic review of 41 studies of assistive technology for people with dementia, including memory aids, which concluded that determining the effectiveness of assistive technology from the available research was difficult because of the small sample sizes of studies, unsophisticated statistical analyses, and the poor functioning of the devices being trialled (Fleming & Sum, 2014).
The findings reported in this review inform several key recommendations for the design and conduct of future studies of electronic memory aids for prospective memory in people with dementia. To ensure widespread uptake and ongoing use of a device, the needs and abilities of people living with dementia need to be well matched with the design and capabilities of the technology (Mason et al., 2012; Mihailidis & Fernie, 2002). Most importantly, the degenerative nature of dementia, which often results in declining memory function over time, needs to be explicitly considered when designing, piloting, and testing devices for this cohort.
As well, studies show that social withdrawal, social isolation, and feelings of loneliness are significant challenges for people with dementia (Dröes et al., 2011; Parikh et al., 2015; Svanström & Sundler, 2015). However, a recent systematic literature review of empirical studies noted that little research has been published which demonstrates the effectiveness of assistive technology for addressing social isolation amongst older adults, including those with dementia (Khosravi et al., 2016). This is another area for particular attention in future studies of electronic aids for prospective memory for people with dementia, such as devices which remind users to attend scheduled social engagements.
In addition to researchers paying closer attention to dementia’s disease trajectory and the implications of this disease for social connectedness, future studies would benefit from two improvements to the design and conduct of the studies themselves. Firstly, more studies should be conducted at a later stage in the development process of devices, which would allow researchers to assess outcomes for users of the devices, such as greater independence and improved quality of life. Admittedly, this may be constrained by the cost of fully developing electronic devices and their associated software prior to testing studies.
Secondly, it would be beneficial for future studies to be conducted with naturally occurring tasks in participants' home environments, rather than in laboratory conditions or in residential care, in order to develop a better understanding of the needs, capacities, and values of people with dementia. Finally, despite the challenges of recruiting and retaining people with dementia to research activities, researchers should aim to obtain larger sample sizes for their studies, so as to achieve statistically significant and generalisable results which can better inform the development of dementia-specific prospective memory aids.
Conclusion
The literature suggests that there are many potential benefits of electronic assistive technology, including prospective memory aids, for people with dementia, their carers and healthcare services. These include reduction in care costs, decrease in caregiver burden, promotion of independence and autonomy, increase in quality of life, and assistance with ageing in place, but these favourable outcomes have not been well established in empirical studies (Bharucha et al., 2009; Cahill et al., 2007; Cash, 2003; Czaja & Rubert, 2002; Czarnuch, Ricciardelli, & Mihailidis, 2016; Dishman & Carrillo, 2007; Orpwood, Gibbs, Adlam, Faulkner, & Meegahawatte, 2005; Rosenberg & Nygard, 2012; Rosenberg et al., 2009).
Overall, while the field shows promise, this review reveals that there is more work to be done to generate robust evidence that effectively informs the design and development of electronic memory aids for prospective memory for people with dementia which significantly improves their quality of life and social connectedness, and enhances their capacity to continue living at home, even as their disease progresses.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
