RFID solutions to support home telecare information flows

• Abstract
• 1. Introduction
• 2. An RFID-based object location system
  • 2.1 Implementation results
• 3. Information flows from the RFID system
• 4. Issues arising
• 5. Discussion
• 6. Acknowledgements
• 7. References

Abstract
Home Telecare is defined as the use of information and communication technologies (ICT) to increase support for people living will illness or disability in their normal activities of daily living, the management of their condition and in their communication with health providers. With an ageing population, New Zealand needs to increase the effort devoted to caring for people with long-term conditions (LTC) in their own homes and Home Telecare is one possible solution. In this paper we discuss the need for increased investment and better management of information for home-based chronic care and disease management. Telecare systems have the potential to enhance the management and integration of home based care through the provision of automated information collected from home telecare or assisted living systems. An example of a novel telecare system, based on the use of radio frequency identification systems is given. This system might be able to assist with finding objects as well as being able to monitor activity in some detail. The potential information flows from this system are described. The information flows that are generated by telecare systems may become important part of the management of long-term and chronic conditions. Issues arising from the use of automated information from home-based care systems are discussed and some potential avenues for improvement and future research are offered.

 

1. Introduction
Throughout the world there has been an increase in the occurrence of long-term conditions (LTC), such as stroke, cancer, diabetes and heart disease and, hence, an increase the importance of delivering effective care efficiently to sufferers. Both for quality of life issues and economic ones, care at home is becoming more important and is being studied intensively.^[1] The demographic shift of the population, from a generally young population to one where the number of workers supporting each elderly person is much smaller, is becoming more visible and many LTCs are associated with increasing age. Data from Statistics New Zealand^[2] based on the "medium" assumption of changes until 2051, estimates that by 2051 the percentage of the population aged 65 years and over will double from 12 percent to 26 percent. A similar scenario exists in the UK where the number of people over the age of 65 has doubled since 1935 and today one-fifth of the population is over 65.^[3] Further, one in every five adults is reported to have some form of disability^[4] with motor and cognitive disability being the most frequent. At the same time, the information flow between health care providers, patients and other stakeholders is being investigated as part of the Health Information Strategy Action Committee (HISAC) process, and being found to be wanting at present, and in need of improvement as part of an action area.^[5]

Thus, a pattern emerges whereby there is a convergence of requirements between the need to assist people to continue to live at home, an increasing need to treat chronic diseases and manage the information required for such processes, and to do so in the context of a holistic health care system. The vast majority of people needing such services are elderly although it should be emphasised that this need is not universal, and does not begin at any specific age.

The New Zealand government bases the assessment of needs for support around a Needs Assessment and Service Co-ordination system (NASC).^[6] The NASC units are based around the country and associated with the district health boards. NASC units do not provide the care, but do identify the level of care needed, and can liaise between providers such as charities, commercial operations and government bodies. The assessment of people’s needs can be complex, but a guide to assessment of older people (and those with existing disabilities) has been published by the New Zealand Guidelines Group.^[7

There is a wide range of technologies used to support people who need assistance in the tasks of daily living. These range from modifications to houses, alarm and fall detection systems, mechanical devices to assist with particular functions (eg, shoe horns), as well as self- and telemonitoring devices such as glucose blood testing kits or blood pressure monitors. Of course, living at home implies less contact with the routine measurement of health status, as may occur in high level care. This may in turn lead to undiagnosed exacerbation or decline in function, which could possibly have been avoided if more information was available to the health professionals with responsibility for the person. In terms of information flow, telemonitoring devices seem an attractive prospect to improve home management of chronic conditions. A recent review^[1] has demonstrated that in the case of hypertension and diabetes, clinical improvement has been shown, but this is not the case for pulmonary and cardiac disease. The review also reported the fact that many published studies were not suitably designed to prove clinical benefit, and that quality of life and economic issues (although vital) were not usually addressed reliably. A more general study of the reasons for adoption or non-adoption of telemedicine services^[8] emphasises the institutional requirements for implementation, adoption, translation and stabilisation of telemedicine so that it becomes normal to use it.

Internationally, countries are investing heavily in telecare. For example, the UK has produced a white paper on community health.^[9]

This paper argues that by combining telemonitoring and assistive technology, there might be benefits both to patients and the health systems that support them. The argument is based on a Radio Frequency Identification (RFID)-based system for object location in the home currently being developed (Section 2).

The potential information flows and some suggestions for the infrastructure to support them are described in Section 3. Section 4 covers issues that may arise from such a system in terms of privacy and ethical issues. Section 5 comprises the discussion.
 

2. An RFID-based object location system
The system being developed has been described elsewhere.^[10] Briefly, the system uses a number of RFID tags located around the user’s home (landmarks), and attached to objects that may be lost. The user is equipped with a tag reading device that allows interaction via screen or voice. The tag reading device operates continuously, recording the detection of landmarks and objects within the range of the detector, also known as the user’s "aura". When the user deposits an object, then it is no longer visible to the system and its location is recorded in reference to the landmarks visited before and after. See figure 1 for potential landmark locations.

Figure 1: Landmark locations

The system differs from other RFID approaches such as LANDMARC,^[11] in that the reader is carried by the user and the tags remain in the same place, rather than the other way round. This approach was taken in order to reduce costs, limit the effect on others sharing the space (eg, spouses) and also to allow the system to be installed in less frequently used areas, or even public spaces. The use of landmark tags, which could have different effective read ranges, allows the spatial resolution of the system to be adjusted as required, so that there can be areas where a difference of 10cm is important (eg, in the drawer area), whereas in a corridor, 5 metres may be the resolution needed. The user in effect builds up a topological rather than a topographic map, so that the relative arrangement of items is more important than some absolute coordinate. The result "Your glasses are between the toaster and the fridge" is more useful to a human than "your glasses are on a bearing of 90 degrees and 7.5 meters away". The latter approach is valuable for a featureless environment such as for avalanche rescue, but not in a complex, obstacle strewn space like the average home.

Essentially, each time an object or landmark RFID tag comes into range of the RFID interrogator (and therefore the user), a database system is updated with a record. This record forms a comprehensive log of where the user has been and what objects they have been in contact with. There is a continuous stream of data coming from the RFID interrogator. However, we sample this data to facilitate the manageability of the database while still providing enough information to be able to use the system. We also delete some of the history on the basis of age and whether new information about that object or landmark has been stored for storage capacity reasons. There is a tremendous opportunity to analyse this data to provide useful information and triggers for care or to help users to learn from the data that is stored in the device. However, simply uploading all of the data collected or making it available to a remote health professional, eg, would be counter productive. Therefore, there is a need with all systems that provide information flows for some sort of intelligence or data management to enable the provision of useful information. Such intelligence would need to relate to the individual case scenario as every patient is unique.
 

2.1.1 Implementation results

A number of experiments have been performed with the system using different combinations of reader and tag. The Phidgit system (Phidgit) uses 125 Khz short range tags. In this experiment tags were mounted on small stalks, and a reader on the hip of the subject, movement through chokepoints – such as doorways (figure 2), chairs and sinks, etc. The data was recorded in a computer in a backpack.

Figure 2: Reader interacting with doorframe tag

 Information about the subject’s movement was derived from a number of these tags mounted around the testing space.

Figure 3: Path generated from recording of tag detection

Because the time of detection was recorded, the velocity of the subject can be derived, and activity estimated. The main experimental issues remain reliability of tag detection and user acceptance.
 

3. Information flows from the RFID system
There are 3 major sets of information flow that can be enabled by the RFID assisted living system:

1. The device acts as an information appliance for the user – gaining information from the environment in terms of the location of objects and presenting this to the user when requested. Other information about the nature of the objects being interrogated, may be provided by external information providers (see 3 below).
2. The summary data of the movement log can provide information to carers at a number of levels. The frequency of landmarks accessed may indicate that the user is active. This information is similar to that provided by such systems as the Liverpool telecare pilot.^[12] There may be key landmarks that are important for activity monitoring (eg, those outside the bedroom, or in the kitchen, or on a lower or upper level of the house) demonstrating that a normal level of daily activity is occurring. In conjunction with the external information providers, the user may wish to allow the carer group access to information about the products they are using – eg, for measuring compliance with self administration of medication.^[13] This information may be useful in the short term to signal an emergency such as a fall, or in the longer term to identify levels of activity, generally. There is also the potential to identify particular types of activity or behaviours, which may be of interest for monitoring changes in activity in response to therapy, or compliance with recommendations for particular behaviours, eg, increased exercise. Such approaches have already been considered for post-marketing surveillance of home automation equipment.^[14]
3. External information providers (such as food and drug manufacturers) may provide information on RFID tags that can assist the user – eg, information on use-by dates, potential issues associated with interactions between products, and information about colour and matching of clothes for those users with impaired vision. Commercial reasons for doing this are already being explored.^[15]

The central aim of our approach is to integrate these information flows with a device that is both useful and usable. It is hoped that such an approach will allow appropriate information to be used to improve patient care.
 

4. Issues arising
The major issues that arise from integration of information flows are concerned with the need for information standards and the privacy of individual users of the system. These are tempered by the need for more affordable and accessible home care, increasing technological sophistication and increases in quality of life expectations.

As the information is traversing many organisational boundaries, current standards for messaging such as HL7 may not be appropriate. Indeed the nature of the information flow is such that messaging approaches may not be required at all, and in fact a status reporting approach should be used. The activity sensor, for example, may send data to a response centre for acute episodes (ie, sudden loss of activity), but provide a summary of trends over a week for the carer. For information coming from information providers, standards are beginning to appear in the market place, in many cases from the providers of barcode standards such as EPCglobal.^[16] The interface between the user and the device, and between different devices, is important and bodies such as the Continua Health Alliance,^[17] are attempting to introduce and enforce standards. The sheer range of potential information sharing devices is sobering – a source for some of them is available at the UK Telecare Knowledge Network,^[18] and websites such as telecare aware.^[19]

The privacy of individual’s information is an important consideration, and information flows have the capacity to compromise this. Work in New Zealand^[20] has shown that a considerable number of people are not happy with seamless transfer of clinical data between health providers. Activity monitoring information may be of use for many different health providers but also be used by relatives and other agencies. It is possible that more information can be derived from combining sources of data, including other assisted living technologies, and home monitoring of blood pressures, etc. The degree to which this information is merged may be an important issue in the future. As a by-product of such systems, information about other occupants of the house may be able to be derived, and this also has implications for privacy. However the RFID-based system has the advantage that it needs to be powered on and carried by the user to work.

Integrated home telecare systems including RFID have recently been described,^[21] and the authors make the point that decentralisation of such systems – that is allowing the user control of information flow – may be important. The privacy issue is complex in a number of ways. Firstly, there is the degree to which users of systems are prepared to sacrifice privacy for convenience – this is an issue which concerns everyone, not just those seeking care at home, and the balance is not certain as sensors proliferate.^[22] Secondly, there is the paradoxical effect of increased privacy that can result from having automatic rather than human helpers; that is being able to live in ones own home without other people whether professional or unpaid carers being constantly present. Privacy issues might be simply overcome by having an opt-in/opt-out facility to give the user more control over information belonging to them. The use of ethical guidelines and codes of practice as well as suitable system encryption would go some way to ensuring patient confidentiality.

There are a number of strong drivers to increase uptake of assisted living systems. The increasing cost of residential care is obviously important. The relative demographic shift of the population, and the increased mobility of the population, has reduced the pool of unpaid carers (eg, relatives of the patient). The raised expectations of quality of life and the increasing technological sophistication of the population also support such a move. These trends are especially clear in Japan,^[23] and government policy there may inform the New Zealand experience.
 

5. Discussion
Information use and sharing is one of the vital tasks of health care systems. However, as increasingly complex care moves into the community, there is a danger that, because of organisational and conceptual barriers, the flow of information will become difficult, inappropriate and less useful. Improved care at home for people with LTCs, is a valid and important aim of health care policy. Technology to support people with LTCs is varied and developing rapidly. The RFID-based system we have been developing has a number of specific roles which we believe could assist in this process. However, unless health information systems that collect information from patients and deliver information to patients and their carers can communicate with home technology hubs, their integration will be problematic.

Current standards for health information transfer may have to be modified and extended to support telecare information flows. It is possible that existing mechanisms such as HL7 may be of use, much as they are currently (eg, in pharmaceutical information transfer). However the transfer of information between carer, health system, patient and other information providers may require new approaches. A system of certification, such as the Health on the Net (HON) code^[24] used for clinical information websites, may be needed to identify valid and useful information providers. Similarly, privacy polices need to not only be in compliance with government standards, but also be simple and comprehensible to ordinary users of the system. Extension of consumer privacy protection policies, such as discussed in,^[25] may assist this process. The range of providers having access to data may be another issue as home monitoring companies, health boards, researchers and NGOs may all wish to share information and pass on information to others including relatives and friends of the patient where required. There are also issues in terms of the sheer amount of data that can be recorded – eg, we would expect that over 100 tags would be used in an average home, and with reasonable degree of movement, activity monitoring could easily yield a reading or more per second, which could lead to tens of thousands of measurements per day. Because tags fail, and reading is not always successful because of range and geometry, processing of this data is not trivial.

Further research and, especially, clinical trials are important in this field. A recent review^[26] pointed out the paucity of good evidence and a vast literature reporting pilots or trials without follow-up. At the same time as understanding any clinical benefits, future work should also focus on the need to gain the greatest benefit from information flows. It is to be hoped that continuing Ministry of Health efforts in this area (including the HISAC and HISO standards development process) will be able to plan a way ahead.

6. Acknowledgement
The authors would like to thank Dr. Richard Curry, University of Portsmouth, and members of the Telecare Knowledge Network for valuable insights into the role of information flows in telecare.
 

7. References
 

1. Pare G, Jaana M, Sicotte C. Systematic review of home telemonitoring for chronic diseases: the evidence base. J Am Med Inform Assoc. 2007 May-Jun;14(3):269-77.
2. Statistics New Zealand. National population projections 2004 (base) – 2051. 2005. http://www2.stats.govt.nz/domino/external/pasfull/pasfull.nsf/7cf46ae26dcb6800cc
   256a62000a2248/4c2567ef00247c6acc256f6b00095c46?OpenDocument. Accessed 28 November 2005.
3. National Statistics UK. Population Trends 2006 Winter; 126: table 1.4 (population: age and sex). www.statistics.gov.uk/downloads/theme_population/PopTrends126.pdf  Accessed 15th October 2007.
4. Corporate Social Responsibility Europe. Disability: some facts http://www.csreurope.org/uploadstore/cms/docs/CSRE_Disability_statistics.pdf
   Accessed 1st October 2007
5. Health Information Strategy Action Committee. Action zone 7 – chronic care and disease management: an initial view. 2007. http://www.moh.govt.nz/moh.nsf/pagescm/532. Accessed 1 July 2007.
6. Ministry of Health. Health needs assessment for New Zealand – an overview and guide 2000. http://www.moh.govt.nz/moh.nsf/pagesmh/941?Open. Accessed 1 July 2007. New Zealand Guidelines Group. Assessment processes for older people. 2003. www.nzgg.org.nz. Accessed 1 July 2007.
7. May C, Harrison R, Finch T, Macfarlane A, Mair F, Wallace P. Understanding the normalization of telemedicine services through qualitative evaluation. J Am Med Inform Assoc. 2003 Nov-Dec;10(6):596-604.
8. UK Department of Health. Our health, our care, our say. 2005, 2 March 2007. http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4127453. Accessed 1 July 2007.
9. Symonds J, Parry D, Briggs J. An RFID-based system for assisted living: challenges and solutions. Paper presented at the ICMCC 2007 event: empowering the patient. Amsterdam 8–10 June 2007.
10. Lionel MN, Yunhao L, Yiu Cho L, et al. LANDMARC: indoor location sensing using active RFID. Wirel Netw. 2004;10(6):701–710.
11. Buckland M, Frost B, Reeves A. Liverpool Telecare Pilot: telecare as an information tool. Inform Prim Care. 2006;14(3):191-6.
12. Loc H, Melody M, Zachary W, et al. A prototype on RFID and sensor networks for elder healthcare: progress report. Paper presented at the Proceeding of the 2005 ACM SIGCOMM workshop on Experimental approaches to wireless network design and analysis, Philadelphia, Pennsylvania, US.
13. Heierman EO III, Cook DJ. Improving home automation by discovering regularly occurring device usage patterns. Paper presented at the Third IEEE International Conference on Data Mining, 2003. ICDM 2003.
14. Want R. Enabling ubiquitous sensing with RFID. Computer. 2004 37(4);84–86.
15. EPCglobal. EPCglobal standards. (2004, 2007). http://www.epcglobalinc.org/standards/. Accessed 1 July 2007.
16. Continua Health Alliance. Continua Health Alliance Home. 2007. from http://www.continuaalliance.org/home. Accessed 1 July 2007.
17. Telecare Knowledge Network. Telecare Knowledge Network. 1 June 2007. http://www.tkn.port.ac.uk/. Accessed 1 July 2007.
18. Hards S. Telecare Worldwide News. (1 July 2007). http://www.telecareaware.com/. Accessed 1 July 2007.
19. Whiddett R, Hunter I, Engelbrecht J, Handy J. Patients’ attitudes towards sharing their health information. Int J Med Inform. 2006 Jul;75(7):530-41.
20. Hsu YL, Yang CC, Tsai TC, Cheng CM, Wu CH. Development of a decentralized telehomecare monitoring system. Telemed J E Health. 2007 Feb;13(1):69-77.
21. Kumagai J, Cherry S. Sensors and sensibility. Spectrum, IEEE. 2004;41(7):22–26, 28.
22. Dethlefs N, Martin B. Japanese technology policy for aged care. Science and Public Policy. 2006;33:47-57.
23. Health on the Net Foundation. HON code of conduct. 2003. http://www.hon.ch/honcode/conduct.html. Accessed 31 December 2003.
24. Miyako O, Koutarou S, Shingo K. RFID privacy issues and technical challenges. Commun ACM. 2005;48(9):66-71.
25. Barlow J, Singh D, Bayer S, Curry R. A systematic review of the benefits of home telecare for frail elderly people and those with long-term conditions. J Telemed Telecare. 2007;13(4):172-9.