JMIR Publications


We are scheduled to perform a server upgrade on Thursday, November 30, 2017 between 4 and 6 PM Eastern Time.

Please refrain from submitting support requests related to server downtime during this window.

Journal of Medical Internet Research

Advertisement

Citing this Article

Right click to copy or hit: ctrl+c (cmd+c on mac)

Published on 28.06.17 in Vol 19, No 6 (2017): June

This paper is in the following e-collection/theme issue:

    Original Paper

    Informing the Design of “Lifestyle Monitoring” Technology for the Detection of Health Deterioration in Long-Term Conditions: A Qualitative Study of People Living With Heart Failure

    1Centre for Assistive Technology and Connected Healthcare, School of Health and Related Research, University of Sheffield, Sheffield, United Kingdom

    2Public Health, School of Health and Related Research, University of Sheffield, Sheffield, United Kingdom

    Corresponding Author:

    Mark S Hawley, BSc (Hons), PhD

    Centre for Assistive Technology and Connected Healthcare

    School of Health and Related Research

    University of Sheffield

    Innovation Centre

    217 Portobello

    Sheffield, S1 4DP

    United Kingdom

    Phone: 44 114 2220682

    Fax:44 114 2220682

    Email:


    ABSTRACT

    Background: Health technologies are being developed to help people living at home manage long-term conditions. One such technology is “lifestyle monitoring” (LM), a telecare technology based on the idea that home activities may be monitored unobtrusively via sensors to give an indication of changes in health-state. However, questions remain about LM technology: how home activities change when participants experience differing health-states; and how sensors might capture clinically important changes to inform timely interventions.

    Objective: The objective of this paper was to report the findings of a study aimed at identifying changes in activity indicative of important changes in health in people with long-term conditions, particularly changes indicative of exacerbation, by exploring the relationship between home activities and health among people with heart failure (HF). We aimed to add to the knowledge base informing the development of home monitoring technologies designed to detect health deterioration in order to facilitate early intervention and avoid hospital admissions.

    Methods: This qualitative study utilized semistructured interviews to explore everyday activities undertaken during the three health-states of HF: normal days, bad days, and exacerbations. Potential recruits were identified by specialist nurses and attendees at an HF support group. The sample was purposively selected to include a range of experience of living with HF.

    Results: The sample comprised a total of 20 people with HF aged 50 years and above, and 11 spouses or partners of the individuals with HF. All resided in Northern England. Participant accounts revealed that home activities are in part shaped by the degree of intrusion from HF symptoms. During an exacerbation, participants undertook activities specifically to ease symptoms, and detailed activity changes were identified. Everyday activity was also influenced by a range of factors other than health.

    Conclusions: The study highlights the importance of careful development of LM technology to identify changes in activities that occur during clinically important changes in health. These detailed activity changes need to be considered by developers of LM sensors, platforms, and algorithms intended to detect early signs of deterioration. Results suggest that for LM to move forward, sensor set-up should be personalized to individual circumstances and targeted at individual health conditions. LM needs to take account of the uncertainties that arise from placing technology within the home, in order to inform sensor set-up and data interpretation. This targeted approach is likely to yield more clinically meaningful data and address some of the ethical issues of remote monitoring.

    J Med Internet Res 2017;19(6):e231

    doi:10.2196/jmir.6931

    KEYWORDS



    Introduction

    Background

    Health services around the world are facing the challenge of providing care for an increasing number of people living with long-term conditions (LTC) within the community [1-3]. The greatest challenges arise from complex or difficult-to-manage conditions—for example, unstable conditions that are subject to exacerbations [4], complex multiple LTCs [5,6], or circumstances that increase vulnerability, such as old age [5] or living alone [7,8]. Health technologies are being developed to help people living at home manage LTCs. One such technology is “Lifestyle Monitoring” (LM), a telecare technology based on the idea that everyday home activities may be indirectly monitored via sensors to give an indication of the changing health state [9-13]. This remote monitoring technology is designed to unobtrusively monitor participants’ home activities over time, with the aim of identifying signs of deterioration as part of the clinical management approaches within a primary or community care setting.

    Activity measures are commonly used to assess medical and social needs in older people (eg, [14,15]) by indirectly testing the effects of disease on the ability to perform everyday activities, such as climbing the stairs or preparing a meal [16]. Currently such tests are mostly carried out either by health care professionals observing the performance of activities or by patients self-reporting their abilities. These tests are carried out either on a single occasion or at infrequent intervals. LM technology has been developed to infer home activities on a continuous basis using sensors sited around the home.

    Reviews of LM research have highlighted deficits in knowledge [17,18]. Key to this are issues around whether and how home activities change during deteriorations in health. Gokalp et al[18] identified three key areas for further investigation in their review: (1) determining the correlation between changes in activities of daily living and well-being or health status (ie, extracting useful information relevant to subjects’ well-being), (2) determining the most effective set of sensors, and (3) the lack of a decision model which explains how changes in monitored activity relate to changes in health, which in turn could inform the actions of health services. This paper focuses on the first two issues: (1) whether and how home activities change during deteriorations in health, and (2) if changes do occur, how these might be captured by sensors sited around the home. The study also informs debate on the third issue raised by the review, relating to the processes of clinical decision-making using information derived from remote monitoring technologies.

    Previous LM studies have described various approaches to remotely monitoring home activities with the aim of detecting deteriorations in health. One common method has been to use sensors to create an electronic version of the activities of daily living (ADL) measure within the home, but this method has so far not provided a practical means of detecting clinically meaningful changes in home activity of patients [17,18]. This study approaches this key issue in a different way. We sought to provide a detailed qualitative exposition of whether and how activities change within the home during the various health states of a long-term condition; and through this lived experience, to inform the design of LM technology.

    Aims

    This paper reports the findings of a study examining everyday activities in people with heart failure (HF) in varying health states in order to identify whether participant activities change in distinctive ways across the health states, and particularly during an exacerbation, in an exemplar long-term condition. We sought to capture details of participants’ activities at home at times when they experience these varying health states to inform sensor set-up for remote monitoring and to potentially capture these clinically important changes within the data stream. The study also explored other influences on everyday activity to understand the degree to which activity was influenced by the symptoms of the condition compared to other factors. The research questions were as follows:

    1. How do everyday activities undertaken at home by people with HF vary according to the health state of the individual carrying them out?
    2. What other factors influence everyday activity in older people?
    3. How might changes in home activity be captured indirectly via sensors sited within the home?

    All research questions were considered through the lens of LM, the remote monitoring technology, and from the perspective of the lived experience of the long-term condition. The focus of the study was HF, but the aims of the study were broader: to add to the to the knowledge base informing the development of home monitoring technologies designed to detect health deterioration in any long-term condition that is subject to exacerbations.

    Within the United Kingdom, approximately 493,000 people are recorded on general practice registers of patients with HF [19]; although, the actual prevalence is likely to be higher. HF is a complex condition with an uncertain trajectory [20-23] and a high symptom burden [24]. The symptoms are known to impact everyday activities (eg, [25-29]) in ways that vary depending on the degree of intrusion from symptoms. For example, Edmonds et al [26] documented changes in activity in response to different degrees of breathlessness. The condition is costly both to the individual and to health services because of the risk of non-elective hospital admission and readmission [30]. It is known that people with HF may delay seeking help during an exacerbation, and therefore, symptoms may deteriorate before health services are alerted [31-33]. One aspect of this delay may arise from difficulties with symptom recognition [32-37]. Early detection of worsening symptoms may potentially halt an exacerbation, allowing the condition to be stabilized via a community health care intervention, thus avoiding the need for hospital admission. This research, therefore, particularly focused on the potential changes in activity leading up to an exacerbation and how these can be detected by sensors.


    Methods

    Participants and Recruitment

    The study participants with HF were recruited in two ways. Participants were either identified by a heart failure specialist nurse and recruited by the research team, or were recruited from a HF support group. All participants lived in and around Barnsley, a town in the north of England.

    The sample was chosen to include a diverse experience of living with HF, and in particular, participants varied in their age, severity of HF, and the length of time since diagnosis (Table 1). All participants were aged 50 years and above. Additional insights were gained from the partners of some of the participants with HF, who, at the participants’ request, sat in and contributed to the interview.

    Study Design

    Participants were interviewed in their own homes over an 8-month period using a semistructured schedule both to ensure that key topics were discussed and to capture a broad range of experience. The key topics were identified by literature searching and included a focus on HF symptoms and activities undertaken during days classified as normal days, bad days, and exacerbations [38,39,25].

    The audio-recorded interviews were typically of one hour duration, with a minimum of 40 minutes and a maximum of two hours. After each interview, field notes were taken to both reflect on the interview and gain insights into the research questions—these notes provided the starting point for data analysis. The interviews were transcribed in full.

    Ethics

    Ethical approval was gained from Leeds (East) Research Ethics Committee (08/H1306/46) and research governance from Barnsley Hospital NHS Foundation Trust. Participants were sent written information about the project at the initial point of recruitment. Key points were reiterated before the interview and participants were informed of their right to stop the interview at any time. Written consent was then taken.

    Theoretical Framework and Analytical Approach

    The project was underpinned by subtle realism: an epistemological stance based on the belief that there are multiple non-contradictory versions of reality and that although accounts differ, they all reflect reality [40]. Researchers sought to cast a critical eye over participants’ accounts of life with HF to find a version “...of whose validity we (were) reasonably confident” ([41], page 50). Researchers were mindful of their own influence on this process and sought, through reflexivity, to identify any erroneous assumptions, attitudes, or beliefs in a process described by O’Cathain [42] as “situating” yourself within the research.

    The interview data were analyzed using template analysis [43-45], a form of thematic analysis where the key tool is the use of a hierarchical template to summarize themes and provide a visual aid to both analysis and data interpretation. The hierarchical structure of the template is used to organize themes in a meaningful way, with key broad concepts captured in higher level themes, and lower level themes providing more detail to improve understanding of specific aspects of the broader themes.

    Data Analysis

    The data were coded by Sarah Hargreaves, with checks made by other researchers to validate choice of codes. A minority of initial themes were informed by the literature (a priori themes); however, these were retained only if the theme was found to be both present within the data and of relevance to the research question. If not, a priori themes were amended to reflect the data or discarded if they were found to be no longer appropriate. Other themes were identified from “features” of participant accounts (perceptions or experiences) relevant to the research questions. Initial theoretical propositions were tested at a validation exercise undertaken at a heart failure support group meeting with 27 attendees. Themes were further consolidated via the process of template development whereby different template structures were explored in order to reach a final version that both represented the themes within the interviews and directly addressed the research questions. Consider, for example, the high level theme “Deciding what to do today,” which arose out of an understanding from the interviews that activities on any given day are influenced by a wide range of factors. These were detailed in lower-level codes, such as the weather, commitments, and activity assessment. The understanding of this theme was consolidated by the validation exercise, which highlighted not only the importance of symptoms as a key influence on daily activity but also individual differences. This process of template development was documented in order to create an audit trail [45].

    The data were initially coded manually and at a later stage imported into NVivo 9 (QSR International) in order to utilize the text retrieval facilities to aid template development and for reporting purposes.

    In this paper, all participants are referred to by pseudonyms.


    Results

    Participant Characteristics

    Table 1 summarizes the characteristics of the interview participants with HF. Participants ranged in age, from younger participants in their 50s to older participants in their 80s, with the majority in their 70s. In addition, many participants suffered from a range of co-morbidities from mild conditions (such as mild arthritis) to potentially life-threatening ones (eg, cancer). The sample included participants who had been diagnosed with HF recently (within the last three years), and those who had been living with HF for longer. Despite efforts to recruit a gender-balanced sample, the sample was predominantly male.

    The partners or spouses of 11 interviewees sat in and contributed to the interviews.

    Table 1. Characteristics of participants with heart failure.
    View this table

    During the analysis an integrative theme was identified [44]. This is a theme which King describes as “undercurrents running through participants’ accounts” (page 460), and is thus pertinent to the interpretation of much of the data. In this study the integrative theme was the degree of intrusion from symptoms.

    The Three Health States of Heart Failure and Impacts on Everyday Activity

    The interviews confirmed that participants recognized the concept of normal and bad days and were able to describe how these two types of days differed both in terms of the degree and type of symptoms experienced, and the sort of activities that they would be likely to undertake. A normal or good day was relatively symptom-free, and this gave participants more energy to undertake everyday activities, which might be within the home. For some, this feeling of well-being inspired participants to undertake activities in the garden or to go out.

    Well on a good day I live like a normal day virtually, don’t I?...Erm if I want to go for a walk, we go for a walk. We go shopping, whatever, visit people.
    [Brian]

    How often these good times occurred, how long they lasted, and the degree of extra energy experienced, all varied depending on the severity of HF. For some, these good times were both infrequent and transient.

    In contrast bad days were characterized by a lack of energy (“a whole system tiredness”), a lethargy, and with it an accompanying feeling of listlessness, irritation, and negativity. During such times participants preferred to stay at home and may have undertaken gentle activities or those requiring very little energy, such as watching the television or using the computer. Again the degree of intrusion from the symptoms varied, with some participants very badly affected by bad days and unable do anything, while others ignored their symptoms and kept going.

    You can have days when you feel really down and you know it just seems like you are useless, you can’t do anything and you don’t have the energy to get a meal or anything
    [Dawn]

    Bad days were regarded as a transitional state to be endured in order to reach better times.

    The third health state was exacerbation. Not all the participants had experienced an exacerbation, but for those who had, it was remembered with horror—as a time when health concerns came to the fore and everything became a struggle. The key symptoms were breathlessness, tiredness, and swelling (oedema), with breathlessness especially reaching frightening levels as participants described struggling for breath.

    ...Just if you were walking up a little incline erm that was something or nowt [nothing] really you had to stop and you would be going [participant acts out doing heavy gasping breaths, followed by a deep lurching breath] and feel as if you couldn’t get enough in.
    [Larry]

    Exacerbations differed from bad days because of the degree of exercise intolerance and dyspnoea, and the perception that the body was out of control, trapped in a “downward spiral.”

    The impact on home activities built up over time, with participants describing how they spent more time at home, and even there, were limited and selective in their activities. A notable aspect of activity within this health state was slowing down, when activities were undertaken very slowly, especially more exerting activities, such as walking and climbing the stairs.

    Erm but just everything was just so slow, so very, very slow.
    [Willow]

    The exacerbation also changed activities at night, with sleep broken by episodes of breathlessness which prompted the participants to undertake activities to ease their breathing, such as sitting on the edge of the bed or going downstairs to sit the night out on the sofa.

    I come down and was watching sport two, three, four o’clock in the morning, and was dozing on the settee. You know awful, awful feeling when you waken in the night-time like that.
    [Larry]

    A number of participants spontaneously described how the first thing that alerted them to the onset of exacerbation was a new or increased difficulty in walking or using the stairs.

    I had been having this increasing breathlessness and then the snow, I couldn’t go further than five yards with a bag and I had to stop myself and think this is ridiculous.
    [Janice]
    It was walking really. I had always been a fast walker and I had to slow down.
    [Willow]

    During an exacerbation of symptoms, walking became an ordeal as the physical effort caused breathlessness. The length of an acceptable walk was scaled down, and the activity was carried out more slowly than normal with stops to breathe. This change was observed both inside the home and outside.

    ...from the top of the stairs into the toilet I was absolutely shattered. From there to the chair I was shattered, from that chair to the end of the settee there [a few steps] and I would have to stand [sound of panting] out of breath completely.
    [Eric]
    ...Just walking to the bottom of the street really slowly and if other people saw me that know me, you should have seen the look on their faces to say what’s the matter with you?
    [Larry]

    As the exacerbation progressed, participants described how activity became pared down to absolute necessities and they avoided climbing the stairs as it became too exhausting and distressing.

    Other Factors Influencing Activity

    The interviews explored the broad experience of living with HF, and the accounts highlighted how participants were chiefly concerned with living their lives in spite of HF.

    Err...I don’t see myself as being bad, poorly, ill. Whatever you want the classification to be. I don’t. Well it is just me, I try not to see that part, I think I am up and running, I am okay.
    [Alistair]

    This stoical attitude presents a challenge to the concept of LM because individuals with HF may not necessarily respond to intrusive symptoms by changing activity. The most extreme example of this came from a participant who, while suffering an exacerbation, ignored his symptoms and attempted to carry on.

    I always remember that the Doctor said to me…he said you are not working Mr Smith are you? Well I have been to the gym this morning, I have been to work all day, I just had a shower and I have run down here. He said you could have collapsed and died, this was the cardiologist.
    [Larry]

    Participant accounts gave an understanding of how everyday activity is influenced by both the individual situation of the participant (eg, severity of HF and co-morbidities, age, personality, gender, lifestyle, attitude to activity), and external factors (eg, housing, local geography, transport availability, weather, demand on time) A good example of this came from Janice, a participant juggling the demands of bringing up her children while living with HF.

    What I think I should do is put an Agatha Christie on the DVD, have someone bring you afternoon tea and put your feet up…But life’s not like that.
    [Janice]

    The interviews illustrate that in real life, the relationship between activity and health is not clear cut, thus LM systems would need to take account of this complexity and individual circumstances.

    Potential Targets for Activity Monitoring to Detect Signs of Exacerbation

    Participant accounts verified that an exacerbation is a distinctively different health state, and they also identified activities that were undertaken in order to ease symptoms. Participants described how symptoms built up over time during the exacerbation, and often they did not seek medical help until the symptoms were intolerable —for some in spite of their close family urging them to seek help.

    The accounts of the experience of activities undertaken during different health states could provide the starting point for developing a platform of LM sensors to monitor home activities linked to an exacerbation of symptoms. Table 2 lists activities which could potentially be monitored as well as insights into how these changes might be monitored. During an exacerbation of symptoms, participants with HF, for example, described both changes to existing activities (eg, reduced speed in walking and climbing the stairs, staying at home more, and being more sedentary) and undertaking new activities in response to distressing symptoms (eg, sitting up at night to aid breathing, perhaps watching television).

    The interviews also highlighted the importance of signs of ill-health relating to appearance, which would require new sensors to be developed. An important sign was swelling, with participants describing the extent of swelling to their body during an exacerbation, as described by Willow:

    My legs were very swollen, if I sat down I needed to get them up and Edward used to have to lift them up because they were just so heavy, I couldn’t lift them up.

    This swelling was reported in feet, legs, the abdomen, and even fingers. More subtle signs noted by the partners of participants with HF were changes in facial appearance including expression, appearance of eyes, blue lips, and alteration in mood and demeanor. In addition, during ill-health, the role of partners changed and they undertook more activities to support their spouse.

    Table 2. Activities that change during an exacerbation and insights into how these changes might be monitored.
    View this table

    Discussion

    Principal Findings

    The basis of LM technology is that the home activities of people at risk of an exacerbation of a long-term condition may be monitored via sensors in order to provide an indication of important changes in health. This study found that a sensor platform could potentially be tailored to identify signs of an exacerbation in HF when the setup is informed by understanding the lived experience of HF, and in particular the impact of intrusion of HF symptoms on everyday activity. We argue that this emphasis on understanding the lived experience of long-term conditions is important both for identifying which health conditions may be appropriate for LM and for tailoring the sensor setup to identify clinically important changes in health. While this study focuses on HF, the approach has a far wider application for the development of approaches to remote monitoring and data analysis in long-term conditions.

    The findings suggest the need for a more personalized and targeted approach to LM to tailor sensor setup to monitor activities which are potential indicators of important changes in health. By exploring home activities during the different health states of HF, we were able to identify activities which should be monitored, and from this, choices could be made about identifying appropriate means of detecting change. This requirement to identify specific activities that are sensitive to changes in symptoms (specific to a health condition) has not been addressed by the LM literature, where descriptions of systems, methods, and the context in which they are being used, tend to be vague [17].

    Although it was known that people with health conditions change their behavior during an exacerbation (eg, during an exacerbation people with HF may sleep upright), detailed description of how these changes occur within the home were lacking [46]. A key finding from this study is the detailed list of changes that occurred during exacerbations (Table 2) and this offers useful information for the developers of sensors and platforms, and to inform the development of algorithms to detect changes in the data collected. This key finding adds to insights gleaned from other studies looking at varied long-term conditions—Clarke, for example [46].

    This targeted approach offers the potential for a reduction in the number of sensors targeted at specific activities, which may provide more clinically meaningful data than a standard broad-brush setup [47]. It may also lead to the development of sensors designed specifically to monitor activities of clinical significance. LM has been criticized for the failure to capture clinically significant data, and for Elwyn et al [48], this arises from the focus on data that is easy to collect rather than data that is indicative of a change in health. It is anticipated that LM data could inform clinical management of patients living within the community, offering a potential safety net, so that primary care services could be alerted to signs of decline in order to intervene early and potentially prevent hospital admission. In our study, participants delayed seeking help for an exacerbation, a finding in keeping with other HF studies [31-33]. A LM service may be of great value to patients, given that it would require no effort from the patient, and may reduce some of the uncertainty and complexity of living with unpredictable conditions.

    An approach focused on collecting data of clinical significance is more ethical, as it is more likely to yield data of value to health professional decision making, leading to interventions of clinical benefit [17,48]. In addition, the creation of a bespoke LM setup for a specific health condition, which also meets the needs of users, is more likely to yield positive benefit for the user and inspire confidence in health professionals. This is a vital step in the wider adoption of LM [49].

    It is likely that LM is more appropriate for some health conditions than others. This study identified some aspects of HF that make it more appropriate for LM. First, the health states within the condition were found to be distinctly different, and in particular there were specific activities found to be undertaken during an exacerbation to ease symptoms, as well as broader changes in activity that occurred (such as staying at home more) during exacerbations. In addition, the exacerbation was found to build up over time, and this would allow sufficient time to identify a change in health and intervene. Participants of the study were able to identify the first signs of an exacerbation, and this could inform the sensor setup. Although taking a bespoke approach to LM design should make it applicable to other long-term conditions, the characteristics of other conditions may make a LM approach less likely to be successful. The buildup to an exacerbation in chronic obstructive pulmonary disease (COPD) is much faster (eg, [50]). A COPD telecare study was unsuccessful in identifying the early stages of an exacerbation due to a lack of early predictors [51], and the authors speculated that the system may have been identifying signs of a bad day rather than an exacerbation.

    Another potential means of enhancing the remote monitoring would be to feed in additional information. The study found that the partners of people with HF were undertaking an informal monitoring role, that is, observing their partner day and night for signs and symptoms that their health was deteriorating. This monitoring included aspects which current sensors would not be able to capture, such as observing changes in the appearance of their partner, their symptoms (observing increased breathlessness, for example), mood, and noting additional activities they were required to undertake. Other studies have noted this monitoring role as well [25]. Such observations could potentially be clinically significant, or highlight a need for closer examination by a health professional [52]. A means of feeding in this information would add an extra valuable dimension to the resources for decision making [53]. There are, however, ethical implications and potential negative impacts on family relationships arising from this approach which would need further exploration.

    Remote monitoring of HF typically focuses on physiological changes [54]. Utilizing technologies that measure physiological changes in parallel with LM provides opportunity to assess the respective accuracy of both data sources and adds an additional depth of information [47]. It would also provide an additional safety net for patients. The first change in activity noted by participants of this study during an exacerbation was a slowing of their walking pace or in climbing the stairs, but they were also aware of increased breathlessness. Unlike physiological measurement technologies, LM technologies require no active engagement from those being monitored, and thus may detect exacerbations in those people unlikely to actively engage with monitoring. The continuous nature of LM may also lead to earlier detection of the beginning of an exacerbation. Future research might focus on how measures of physiological change and activity vary across the timeline of an exacerbation within the complexity of the lived experience of HF.

    Complexity arises from placing technology within the home, where the monitoring interacts with the individual circumstances in which a person lives and the external factors impacting on everyday life. The health state exists not as a separate entity, which directs home activity, but within a complex environment, where factors interact and there may be conflicting demands. This presents a challenge to LM because an individual may feel ill but not attend to those feelings by resting or staying at home, as their lifestyle has other demands. It is well known that people do not respond uniformly to the challenges of ill health [55-58]; and that people with health issues may prefer to focus on living their lives, rather than the health condition [55,58,59]. This means that there will be a level of uncertainty in the data recorded by LM sensors. However we argue that by understanding more about an individual and their attitudes, priorities, circumstances, and routine, this would provide a context within which to both tailor sensor setup to fit in with individual circumstances and behaviors, and also a means of interpreting the LM data in order to reduce the margin of error [60]. We argue that this understanding is important in order to establish a link between the sensor data and the person being monitored. This would require a degree of personalization of sensor setup and interpretation, which would involve additional costs [61]; it may also involve the development of new sensors. This requirement for technologies to be designed “in the wild,” taking account of how real people live in their homes is an important element in designing technologies to fit needs [62].

    Limitations

    The participants recruited to this study were varied in terms of age and HF severity, and therefore may not be representative of the typical elderly HF patient cared for by general practice [63]. This study focused entirely on HF, but for some the situation is complicated by co-morbidities [30], and the presence of these is likely to make LM more challenging. The small sample size is a limitation of the study, as it is unlikely that we have captured all aspects of behavior within the home in individuals with HF.

    There was an imbalance within the sample, with a greater proportion of male participants and few participants who lived alone. In addition, the study participants were all from a similar geographical area within Northern England and were from the same ethnic grouping (white British), and this may have in part influenced the way that participants responded to the challenges of living with a long-term condition. Certainly participants were stoical in their approach to living with HF, which may reflect the culture of an area with a history of coal-mining and steel-working. This stoical approach to living with a long term condition has been noted by another study based in Northern England [64].

    It is acknowledged that there is a risk that the presence of partners may have influenced the interviews (eg, [65-67]). However, we argue that the presence of partners was positive as they were able to provide vivid descriptions of the impact of HF on activity and to act as a counterbalance to the instinct of many of the participants with HF to underplay symptoms.

    Conclusions

    This study shows the importance of understanding of the lived experience of long-term conditions in order to potentially identify which conditions are suitable for LM and of detailed information about how home activities change when health deteriorates. This detailed information is the key to designing a sensor setup tailored to both specific health conditions and individual circumstances and to inform data analysis and interpretation. By this means it is anticipated that LM can provide timely, clinically meaningful data that will contribute to the management of long-term conditions of people living in the community.

    Acknowledgments

    The Authors wish to thank Professor Nigel King from the University of Huddersfield for his expert advice on template analysis. Dr Hargreaves was funded by Barnsley Hospital NHS Foundation Trust.

    Conflicts of Interest

    None declared.

    References

    1. Campbell D. The Guardian. 2014 Jan 03. NHS could be 'overwhelmed' by people with long-term medical conditions   URL: https://www.theguardian.com/society/2014/jan/03/nhs-overwhelmed-long-term-medical-conditions [accessed 2017-06-05] [WebCite Cache]
    2. Christensen K, Doblhammer G, Rau R, Vaupel JW. Ageing populations: the challenges ahead. The Lancet 2009 Oct;374(9696):1196-1208. [CrossRef]
    3. World Health Organization. WHO. Geneva: World Health Organization; 2011. Global status report on noncommunicable diseases 2010: description of the global burden of NCDs, their risk factors and determinants   URL: http://www.who.int/nmh/publications/ncd_report2010/en/ [accessed 2017-06-05] [WebCite Cache]
    4. Department of Health, Social Services and Public Services. Health-ni. 2012. Living with long term conditions policy framework   URL: https://www.health-ni.gov.uk/publications/living-long-term-conditions-policy-framework [accessed 2017-06-05] [WebCite Cache]
    5. Department of Health. Gov.uk.: Department of Health; 2012. Long term conditions compendium of information, third edition   URL: https:/​/www.​gov.uk/​government/​publications/​long-term-conditions-compendium-of-information-third-edition [accessed 2017-06-05] [WebCite Cache]
    6. Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. The Lancet 2012 Jul;380(9836):37-43. [CrossRef]
    7. Kharicha K, Iliffe S, Harari D, Swift C, Gillmann G, Stuck AE. Health risk appraisal in older people 1: are older people living alone an “at-risk” group? Br J Gen Pract 2007;57(537):271-276.
    8. Haslbeck JW, McCorkle R, Schaeffer D. Chronic illness self-management while living alone in later life: a systematic integrative review. Res Aging 2012 Jan 10;34(5):507-547. [CrossRef]
    9. Celler BG, Hesketh T, Earnshaw W, Ilsar ED. An instrumental system for the remote monitoring of changes in functional health status of the elderly at home. 1994 Presented at: Proceedings of the 16th Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Engineering Advances: New Opportunities for Biomedical Engineers; 1994; Baltimore, MD p. 908-909.
    10. Celler BG, Lovell NH, Hesketh T, Ilsar ED, Earnshaw W, Betbeder-Matibet L. Remote home monitoring of health status of the elderly. Medinfo 1995;8 Pt 1:615-619. [Medline]
    11. Kaushik AR, Celler BG. Characterization of PIR detector for monitoring occupancy patterns and functional health status of elderly people living alone at home. Technol Health Care 2007;15(4):273-288. [Medline]
    12. Kaushik AR, Lovell NH, Celler BG. Evaluation of PIR detector characteristics for monitoring occupancy patterns of elderly people living alone at home. 2007 Presented at: Annual International Conference of the IEEE Engineering in Medicine and Biology Society; 2007; Lyon, France p. 3802-3805.
    13. Ní Scanaill SC, Carew S, Barralon P, Noury N, Lyons D, Lyons GM. A review of approaches to mobility telemonitoring of the elderly in their living environment. Ann Biomed Eng 2006;34(4):547-563. [Medline]
    14. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged. The index of ADL: a standardized measure of biological and psychosocial function. JAMA 1963;185:914-919.
    15. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist 1969;9(3):179-186. [Medline]
    16. Bowling A. Measuring Health: A Review of Quality of Life Measurement Scales. Buckingham: Open University Press; 1997.
    17. Brownsell S, Bradley D, Blackburn S, Cardinaux F, Hawley MS. A systematic review of lifestyle monitoring technologies. J Telemed Telecare 2011 Apr;17(4):185-189. [CrossRef] [Medline]
    18. Gokalp H, Clarke M. Monitoring activities of daily living of the elderly and the potential for its use in telecare and telehealth: a review. Telemed J E Health 2013 Dec;19(12):910-923. [CrossRef] [Medline]
    19. Townsend N, Bhatnagar P, Wilkins E, Wickramasinghe K, Rayner K, British Heart Foundation Centre on Population Approaches for Non-Communicable Disease Prevention, Nuffield Department of Population Health, University of Oxford. Cardiovascular disease statistics.: British Heart Foundation; 2015.   URL: https://www.bhf.org.uk/publications/statistics/cvd-stats-2015 [accessed 2016-10-14] [WebCite Cache]
    20. Jones AM, O'Connell JE, Gray CS. Living and dying with congestive heart failure: addressing the needs of older congestive heart failure patients. Age Ageing 2003;32(6):566-568. [Medline]
    21. Murray SA, Kendall M, Boyd K, Sheikh A. Illness trajectories and palliative care. BMJ 2005 Apr 30;330(7498):1007-1011 [FREE Full text] [CrossRef] [Medline]
    22. Witham MD, Argo IS, Johnston DW, Struthers AD, McMurdo ME. Predictors of exercise capacity and everyday activity in older heart failure patients. Eur J Heart Fail 2006 Mar;8(2):203-207 [FREE Full text] [CrossRef] [Medline]
    23. Yu DS, Lee DT, Kwong AN, Thompson DR, Woo J. Living with chronic heart failure: a review of qualitative studies of older people. J Adv Nurs 2008 Mar;61(5):474-483. [CrossRef] [Medline]
    24. Zambroski CH, Moser DK, Bhat G, Ziegler C. Impact of symptom prevalence and symptom burden on quality of life in patients with heart failure. Eur J Cardiovasc Nurs 2005 Sep;4(3):198-206. [CrossRef] [Medline]
    25. Clark AM, Reid ME, Morrison CE, Capewell S, Murdoch DL, McMurray JJ. The complex nature of informal care in home-based heart failure management. J Adv Nurs 2008 Feb;61(4):373-383. [CrossRef] [Medline]
    26. Edmonds PM, Rogers A, Addington-Hall JM, McCoy A, Coats AJ, Gibbs JS. Patient descriptions of breathlessness in heart failure. Int J Cardiol 2005 Jan;98(1):61-66. [CrossRef] [Medline]
    27. Evangelista LS, Dracup K, Doering LV. Treatment-seeking delays in heart failure patients. J Heart Lung Transplant 2000 Oct;19(10):932-938. [Medline]
    28. Jeon Y, Kraus SG, Jowsey T, Glasgow NJ. The experience of living with chronic heart failure: a narrative review of qualitative studies. BMC Health Serv Res 2010 Mar 24;10:77 [FREE Full text] [CrossRef] [Medline]
    29. Pattenden JF, Roberts H, Lewin R J P. Living with heart failure; patient and carer perspectives. Eur J Cardiovasc Nurs 2007 Dec;6(4):273-279. [CrossRef] [Medline]
    30. Donkor A, Cleland J, McDonagh T, Hardman S, Aminat SA, the National Heart Failure Audit Steering Group Committee. UCL. 2015. National heart failure audit: April 2014 to March 2015   URL: https://www.ucl.ac.uk/nicor/audits/heartfailure/documents/annualreports/heartfailurepublication14_15 [accessed 2017-06-05] [WebCite Cache]
    31. Clark AM, Freydberg CN, McAlister FA, Tsuyuki RT, Armstrong PW, Strain LA. Patient and informal caregivers' knowledge of heart failure: necessary but insufficient for effective self-care. Eur J Heart Fail 2009 Jun;11(6):617-621 [FREE Full text] [CrossRef] [Medline]
    32. Horowitz CR, Rein SB, Leventhal H. A story of maladies, misconceptions and mishaps: effective management of heart failure. Soc Sci Med 2004;58(3):631-643. [Medline]
    33. Riegel B, Dickson VV, Cameron J, Johnson JC, Bunker S, Page K, et al. Symptom recognition in elders with heart failure. J Nurs Scholarsh 2010 Mar;42(1):92-100. [CrossRef] [Medline]
    34. Carlson B, Riegel B, Moser DK. Self-care abilities of patients with heart failure. Heart Lung 2001;30(5):351-359. [CrossRef] [Medline]
    35. Gallagher R. Self management, symptom monitoring and associated factors in people with heart failure living in the community. Eur J Cardiovasc Nurs 2010 Sep;9(3):153-160. [CrossRef] [Medline]
    36. Giamouzis G, Kalogeropoulos A, Georgiopoulou V, Laskar S, Smith AL, Dunbar S, et al. Hospitalization epidemic in patients with heart failure: risk factors, risk prediction, knowledge gaps, and future directions. J Card Fail 2011 Jan;17(1):54-75. [CrossRef] [Medline]
    37. National Collaborating Centre for Chronic Conditions (UK). NICE. London: Royal College of Physicians (UK); 2010. Chronic heart failure: national clinical guideline for diagnosis and management in primary and secondary care   URL: https://www.nice.org.uk/guidance/cg108/evidence/full-guideline-pdf-136060525 [accessed 2017-06-05] [WebCite Cache]
    38. British Heart Foundation. London: British heart foundation   URL: https://www.bhf.org.uk/publications/heart-conditions/an-everyday-guide-to-living-with-heart-failure [accessed 2016-10-14] [WebCite Cache]
    39. Charmaz K. Good Days, Bad Days: The Self in Chronic Illness and Time. New Brunswick, NJ: Rutgers University Press; 1991.
    40. Murphy E, Dingwall R. The ethics of ethnography. In: Atkinson P, Coffey A, Delamont S, Lofland J, Lofland L, editors. Handbook of ethnography. London: Sage Publications; 2007:339-351.
    41. Hammersley M. What's Wrong With Ethnography. Routledge: London; 1992.
    42. O'Cathain A. Reporting Mixed Methods Projects. In: Andrew S, Halcomb EJ, editors. Mixed methods for nursing and the health sciences. Oxford: Wiley-Blackwell; 2009:135-138.
    43. Crabtree BJ, Miller WL. Using codes and code manuals. In: Crabtree BL, Miller WL, editors. Doing Qualitative Research. Thousand Oakes, CA: Sage Publications; 2012:163-178.
    44. King N. Doing template analysis. In: Symon G, Cassell C, editors. Qualitative Organizational Research: Core Methods and Current Challenges. London: Sage; 2012:426-450.
    45. King N. School of Human Health Sciences, University of Huddersfield. Template analysis   URL: http://www-old.hud.ac.uk/hhs/research/template-analysis/ [WebCite Cache]
    46. Clarke M. The need for an integrated approach to remote monitoring of physiological data and activity data. J Telemed Telecare 2014 Apr;20(3):159-160. [CrossRef] [Medline]
    47. Glascock AP, Kutzik DM. Embrace the Chaos, It's Not Noise: lessons learned from non-traditional environments. IJMHCI 2009;1(2):22-36. [CrossRef]
    48. Elwyn G, Hardisty AR, Peirce SC, May C, Evans R, Robinson DK, et al. Detecting deterioration in patients with chronic disease using telemonitoring: navigating the 'trough of disillusionment'. J Eval Clin Pract 2012 Aug;18(4):896-903. [CrossRef] [Medline]
    49. Barrett D, Thorpe J, Goodwin N. Examining perspectives on telecare: factors influencing adoption, implementation, and usage. Smart Homecare Technol Telehealth 2015;3:1-8. [CrossRef]
    50. Hardisty AR, Peirce SC, Preece A, Bolton CE, Conley EC, Gray WA, et al. Bridging two translation gaps: a new informatics research agenda for telemonitoring of chronic disease. Int J Med Inform 2011 Oct;80(10):734-744. [CrossRef] [Medline]
    51. Pinnock H, Hanley J, McCloughan L, Todd A, Krishan A, Lewis S, et al. Effectiveness of telemonitoring integrated into existing clinical services on hospital admission for exacerbation of chronic obstructive pulmonary disease: researcher blind, multicentre, randomised controlled trial. BMJ 2013 Oct 17;347:f6070. [CrossRef]
    52. Royal College of Nursing. RCNHCA. First steps for healthcare assistants   URL: http://rcnhca.org.uk/clinical-skills/observation/ [WebCite Cache]
    53. Brownsell S, Bradley D, Cardinaux F, Hawley MS. Developing a systems and informatics based approach to lifestyle monitoring within eHealth: part II analysis and interpretation. 2011 Jul Presented at: Healthcare Informatics, Imaging and Systems Biology (HISB), First IEEE International Conference on th July; 2011; San Jose, CA p. 26-29. [CrossRef]
    54. Bashi N, Karunanithi M, Fatehi F, Ding H, Walters D. Remote monitoring of patients with heart failure: an overview of systematic reviews. J Med Internet Res 2017;19(1):e18. [CrossRef]
    55. Carel H. Illness: The Cry of the Flesh. Durham: Acumen Publishing; 2008.
    56. Donovan JL, Blake DR, Fleming WG. The patient is not a blank sheet: lay beliefs and their relevance to patient education. Br J Rheumatol 1989 Feb;28(1):58-61. [Medline]
    57. Furze G, Donnison J, Lewin R. Clinician's Guide to Chronic Disease Management of Long Term Conditions: A Cognitive Behavioural Approach. Keswick: M&K Publishing; 2008.
    58. Mock C, MacKenzie E, Jurkovich G, Burgess A, Cushing B, deLateur B, et al. Determinants of disability after lower extremity fracture. J Trauma 2000 Dec;49(6):1002-1011. [Medline]
    59. Charmaz K. The body, identity, and self: adapting to impairment. Sociol Q 1995;36(4):657-680.
    60. Hanson J, Osipovic D, Percival J. Making sense of sensors: older people’s and professional caregivers’ attitudes towards telecare. In: Loader B, Hardy M, Keeble L, editors. Digital Welfare for the Third Age: health and social care informatics for older people. Abindgon, UK and New York, NY: Routledge; 2009:91-111.
    61. Brownsell S, Blackburn S, Hawley M. User requirements for an ICT-based system to provide care, support and information access for older people in the community. J Assist Technol 2012;6(1):5-23. [CrossRef]
    62. Wherton J, Sugarhood P, Procter R, Rouncefield M, Dewsbury G, Hinder S, et al. Designing assisted living technologies 'in the wild': preliminary experiences with cultural probe methodology. BMC Med Res Methodol 2012 Dec 20;12:188 [FREE Full text] [CrossRef] [Medline]
    63. Mosterd A, Hoes AW. Clinical epidemiology of heart failure. Heart 2007 Sep;93(9):1137-1146 [FREE Full text] [CrossRef] [Medline]
    64. King N, Carroll C, Newton P, Dornan T. “You can't cure it so you have to endure it”: the experience of adaptation to diabetic renal disease. Qual Health Res 2002 Mar;12(3):329-346. [CrossRef] [Medline]
    65. Boeije HR. And then there was three: self-presentation styles and the presence of the partner as a third person in the Interview. Field Methods 2004;16(1):3-22. [CrossRef]
    66. Kendall M, Murray SA, Carduff E, Worth A, Harris F, Lloyd A, et al. Use of multiperspective qualitative interviews to understand patients' and carers' beliefs, experiences, and needs. BMJ 2009 Oct 14;339:b4122. [CrossRef]
    67. Reuband KH. On third persons in the interview situation and their impact on responses. Int J Public Opin Res 1992;4(3):269-274. [CrossRef]


    Abbreviations

    ADL: activities of daily living
    COPD: chronic obstructive pulmonary disease
    HF: heart failure
    LM: lifestyle monitoring
    LTC: long-term conditions


    Edited by G Eysenbach; submitted 01.11.16; peer-reviewed by M Clarke, K Antypas; comments to author 29.01.17; revised version received 21.03.17; accepted 28.04.17; published 28.06.17

    ©Sarah Hargreaves, Mark S Hawley, Annette Haywood, Pamela M Enderby. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 28.06.2017.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.