eCAALYX builds on the experience gained in a previous and closely related EU-funded project, CAALYX (Complete Ambient Assisted Living Experiment, 2007-2008; [23, 30, 33]). The original CAALYX prototype similarly used a smartphone and one of the prototype's commercial exploitation barriers identified at that time was the relatively high cost of acquiring an Internet- and GPS-enabled smartphone with sufficient computing power (and battery life) to run the CAALYX mobile app (plus the subscription costs to a suitable mobile phone data plan), particularly given the generally low income of target users (older pensioners [34]). However, recent surveys of mobile phone uptake and penetration in the UK and other developing countries show that this affordability barrier might be gradually improving on the medium to long run, especially for the "younger" older generations (55-70) and as smartphone prices and data plan costs continue to drop.
A closer look at ownership of smartphones
Smartphone ownership in the developed world is rapidly increasing. ChangeWave Research [35] report a (presumably US) survey in December 2009 with 42% saying they own a smartphone (rising from 15% in October 2006) (but details of the survey methods are not freely available). Estimates from Ofcom research [36] suggest the figure is less, namely that the USA has 36% and the UK has 37% with Internet access via a mobile phone. Ofcom also cites ComScore's MobiLens survey [37] estimating only 18% of people with smartphone subscriptions in the UK compared to 26% in Italy and 21% in Spain. Ofcom reports that there is a good deal of consumer confusion over the capabilities of their phone but that 43% claim their phone can be used to access apps, download email, and surf the Web. If 81% have a mobile phone (see below) this would imply a third of the UK population with smartphones. We might estimate therefore that between one fifth and one third of the UK population now has a smartphone.
What are the predictions for smartphone ownership to reach over 90% of the population?
We can perhaps extrapolate from ownership of any mobile phone. UK's National Statistics [38] report the growth in mobile phone ownership since 2001/02, increasing from 65% to 81% in 2009. Ownership varies by income group. Only 67% of households in the lowest income decile group reported mobile phone ownership in 2009, compared with 92% in the highest income decile group. (Recent estimates of mobile phone use in the USA are 85% [39], so there is no significant difference between UK and USA (given the differences in methods used and error in these estimates).)
Age is probably the major 'digital divide'. The Continuous Household Survey in Northern Ireland [40] provides a good indication of how mobile phone ownership is increasing. It puts current ownership at 89% in Northern Ireland. It is not clear whether this higher figure compared to the British 81% from National Statistics is 'real' or due to methodological differences. Nevertheless, the pattern of older consumers lagging behind the rest of the population (Figure 4) is likely also to be the case in England and the rest of the UK. However, if the current trajectory continues, over 95% of the over 60 s will have a mobile phone in five years.
Taking these two trends together we might estimate that 80-90% of the UK population will have a smartphone within 10 years. This might not be quite as rapid as some might suggest, but, in the context of other social changes brought about by the development of technology, is a massive change.
What impact will this have on healthcare?
Jane Sarasohn-Kahn reports "How smartphones are changing healthcare for consumers and providers" [41]. Probably the quickest change is in the use of smartphones by professionals. The slightly slower, but possibly bigger, impact may be in the use of smartphones by lay consumers. Producers of apps certainly think there is a market, as by February 2010, some 4000 apps were available within the Apple App Store aimed at patient end-users, and Gartner named mobile health as one of its top ten applications for 2012 [42].
Considering that many apps now, especially those intended for chronic disease management, rely on the presence of an active Internet connection on the smartphone in order to function as intended, a question immediately arises about current Internet uptake levels among people with chronic diseases.
Do people with chronic disease go online and can Internet-enabled smartphones make a difference?
Fox and Purcell [43] report that in the USA, adults living with chronic disease are disproportionately offline. 81% of adults reporting no chronic diseases go online compared to 62% of adults living with one or more chronic disease. And there is a marked 'dose response', so 68% of adults reporting one chronic disease go online compared to only 52% of adults living with two or more chronic diseases. The difference is even more marked when comparing use of text messages with only 23% of those with two or more chronic conditions compared to 60% of those with no conditions using a mobile to send text messages [43].
There are now many health promoting Internet interventions [44] but they will, by definition, only be used by those who have already reached a decision to try to change their behaviour. Although many report successful behaviour change in those who continue to use such interventions, attrition is usually very high, and we can assume that those who drop out of using the online intervention have also dropped out of changing their behaviour. It may be that smartphone technology, by its mobility and location awareness, may be able to achieve lower attrition rates, but given the demographics of current users it is likely to be for behaviour change amongst a relatively young and healthy population. Sarasohn-Kahn [41] had argued that medication adherence is a problem amongst patients with chronic conditions and suggested that technology can play an important role. However, there is no strong evidence yet on the effectiveness of dispensing devices, and like behaviour change, this is likely only to benefit those willing to use such reminders and who are already smartphone users, unless apps running in specialised devices can be developed.
There are now hundreds of smartphone apps focusing on wellness, fitness, and nutrition. Mobile and home monitoring can be carried out with body-worn and ambient sensors communicating with smartphones (as found, for example, in eCAALYX), including accelerometers measuring motion and gait, infrared detectors measuring body temperature and heart rate, and glucometers measuring blood glucose (some sensors may also be built-into future smartphones). However, Sarasohn-Kahn [41] describes how MedApps started as a mobile phone app, but given the low user base amongst chronically ill and older people they re-engineered it as a wireless handheld device, the HealthPal. Users use their 'usual' devices such glucometers, spirometers, pulse oximeters, and scales, and the Bluetooth enabled Healthpal communicates results. Another rather obvious consideration in smartphone use for people who are (chronically) ill, is that most such people are at home, and unlikely to be 'out and about' using their mobile phone. In that circumstance they are more likely, with current device usage, to use a Wi-Fi enabled laptop (or iPad/Android Touch Tablet) from their bed (although the use of 3G mobile Internet connections indoors is also on the rise on different devices ranging from smartphones to notebooks). However, smartphones, especially newer ones with larger (touch) screens, are starting to replace conventional desktop and Wi-Fi access (and some smartphones are both Wi-Fi and 3G mobile Internet enabled). In the UK, 29% of Internet users use their mobile to access the Internet at home, second only to those in Japan at 43% [36], and mobiles are rapidly replacing landlines in many homes. In the UK, 13% of all homes used mobiles as their sole form of telephony [36], but more than half of Americans aged 25-29 now live in households with mobile phones but no traditional landline telephones [1].
At the moment, Sarasohn-Kahn [41] cites Eising of Mayo clinic mentioning their research that people use mobiles for 'action-oriented' information, and how they are not going to do in-depth research by mobile. So smartphones will be used by people who are not acutely ill but who maybe want to find some location based information - such as the location and hours of a pharmacist - while on the move. Whereas more detailed information, or communication with others may take place in the home. But as more people get smartphones and use them as their sole means of communication, this may change.
What other barriers to smartphone use in health does Sarasohn-Kahn [41] see? She is concerned that too much app development is carried out by technologists without the involvement of patients. She notes the problems of apps recommending particular products and also notes that in the USA if an app includes a sensor then the FDA (US Food and Drug Administration) may monitor it as a medical device. She thinks that the challenges to continued rapid smartphone growth include finding the right business model and privacy issues. (For a brief discussion of the latter privacy issues and some workarounds adopted in CAALYX/eCAALYX, interested readers may refer to [23] and [45]).
Some limitations of mobile phone applications
Notwithstanding all of the benefits we should be aware that the use of the mobile phone in healthcare and clinical practice is not without its problems and limitations. In comparison to laptop computers, the small internal storage capacity, processing power and screen size of the mobile phone often requires apps that are running to be in reduced format [25]. However, the use of cloud computing resources which are external to the mobile device may obviate restrictive processing speeds and memory requirements in the future [25]. Never the less, mobile phones are smaller, more portable and less obtrusive than laptops, so it could be argued that this is a reasonable trade off. Although much mobile phone communication is now conducted using text, voice communication still necessitates the securing of space within which vocal communication can be made in private [46]. The consideration of such a constraint may be vital to maintain the confidentiality of patient information if used in public spaces. Other factors such as loss or theft of devices may impact upon the security of confidential digital health records or data held on mobile phones. The security of patient data held on mobile phones has been a concern for some time [47], while some studies warn of the security risks of using mobile instant messaging in healthcare [48].
Patient attitudes and perceptions
When used as a method for monitoring the health status of remote patients, mobile phones should be applied only after due consideration to patient perceptions and feelings. One study in Italy revealed that patients with implantable cardioverter-defibrillators for cardiac resynchronisation therapy welcomed the use of mobile technologies for remote monitoring, but did not want them to replace their personal contact with health workers [49]. Other studies using mobile technology for remote monitoring of health conditions found similar results [50, 51]. The Canadian study by Seto et al. [51] also advised that mobile phone-based remote monitoring will not be suitable for all patients. Those for example who suffer from poor manual dexterity, failing vision or a predisposition to high levels of anxiety may not be able to operate the remote mobile monitoring tools. These results are supported by similar findings from a study of older patients with disabilities in the USA [52]. It should also be acknowledged that prosaic issues such as remembering to recharge a device and the simple maintenance of equipment within a patient's home may be problematic.
Kurniawan [53], Lorenz and Oppermann [54], and Gao and Koronios [55] provide detailed discussions of mobile app needs of older people, covering aspects such as ergonomics and user interface issues (e.g., memory aids, visual aids, haptic aids, features to minimise user error, and safety features), as well as the most commonly desired smartphone app functionalities in this age group. However, senior citizens of tomorrow will include the young and middle aged of today, who are more familiar with, and reliant on computers, smartphones and the Internet than previous generations, and are increasingly well-versed in using these technologies on a daily basis for study, work and leisure. This might partially contribute (in the long run) to solving some of the smartphone app usability and learnability issues, which current generations of older people are facing.
Which platform to support? A developer's dilemma
According to a recent MobiHealthNews report published in November 2010 [56], from February to September of the same year, Google's Android smartphone saw a 156.6% increase in the number of available health-related apps, compared with a 66.6% increase in Apple's health-related apps. The number of health-related apps in BlackBerry's App World increased by 141.4% over the same period. However, Apple is still leading in terms of the absolute total number of health-related apps available on any platform. As of September 2010, Apple's App Store offered the highest number of health-related apps at 7,136, followed by Google Android at 1,296 and BlackBerry at 338. These figures represent a healthy 78% increase in the number of health-related apps on these three platforms combined since February 2010. The MobiHealthNews report did not cover the latest Microsoft Windows Phone 7 platform as this was only officially launched during October/November 2010; older versions of the Windows platform (Windows Mobile 6.x) had less than a 3% market share of worldwide 2010 Q3 smartphone sales (vs. 25.5% for Android, 16.7% for Apple iOS and 14.8% for Research In Motion--RIM BlackBerry) [42], and were probably not considered in the MobiHealthNews report for this reason. However, the emerging Windows Phone 7 platform is rapidly gathering momentum and support by major smartphone providers [57], with a growing number of health-related apps already available from Windows Phone 7 Marketplace for the new handsets (Figure 1 - [3]). It remains to be seen how the emerging Windows Phone 7 platform will fare against the well-established competitor smartphone players and their planned updates in 2011/2012.
Other constraints
Desk-based health researchers who rely on the telephone to gather their data are faced with a growing problem. Increasingly, respondents are replacing their landlines with mobile telephones, and in so doing, they create a problem for the researcher. Legislation in some countries requires researchers to manually dial mobile phone numbers, thereby incurring significantly more cost in both time spent on calling and in call costs. Further, calling a mobile phone on some tariffs may use up respondents' air time, and there may therefore be an ethical onus on the researcher to reimburse the costs incurred [58].