‘Somebody there to watch over you’: the role of the family in everyday and emergency diabetes care


  • Kathleen L. Gunn MSc,

    1. Royal College of Nursing Research Institute, School of Health and Social Studies, University of Warwick, Coventry, UK
    Search for more papers by this author
  • Kate Seers PhD RN,

    1. Royal College of Nursing Research Institute, School of Health and Social Studies, University of Warwick, Coventry, UK
    Search for more papers by this author
  • Natasha Posner PhD MA,

    1. Royal College of Nursing Research Institute, School of Health and Social Studies, University of Warwick, Coventry, UK
    2. Present address: School of Advanced Study, Senate House (South Block), University of London, Malet St., London WC1E 7HU, UK
    Search for more papers by this author
  • Vivien Coates PhD MPhil RGN

    1. University of Ulster/Western Health and Social Care Trust, Institute of Nursing Research, University of Ulster, Coleraine, UK
    Search for more papers by this author

Kathleen L. Gunn
Royal College of Nursing Research Institute
School of Health and Social Studies
University of Warwick
Coventry CV4 7AL, UK
E-mail: kathleen.gunn@warwick.ac.uk


This paper reports on the role of family members in everyday diabetes self-care and in diabetic crises. It is based on qualitative data drawn from 45 semi-structured interviews with a wide range of people with an established diagnosis of Type 1 or Type 2 diabetes, who were admitted to hospital for urgent or emergency treatment in connection with their diabetes. The interviews were carried out in two contrasting sites in the United Kingdom in 2009–2010, transcribed and analysed thematically with particular reference to framework analysis methods. We found that family involvement in self-care was common, and the role of family and friends was especially important when the person with diabetes needed urgent help. We comment on the diversity of family members who assisted regularly or dealt with crises, the importance of taking account of the complexities of family life, including reciprocal care, and the particular problems faced by people without family support. Finally, we make recommendations for further research and for improvements in existing services.

What is known about this topic

  •  Self-care is key to the control of diabetes and the prevention of complications.
  •  Diabetes self-care may be supported or otherwise influenced by family members.

What this paper adds

  •  The role of family members is critically important, both day-to-day and in crisis, and it is vital that training and care plans reflect this.
  •  With the increasing number of people diagnosed with diabetes, the increase in co-morbidities and changing demographics in which there are more frail, elderly people in society, this need for family help is likely to grow.
  •  Constant vigilance is a draining activity and the toll that it takes on family members, including young carers, needs to be taken into account.
  •  People who do not have relatives to look out for them need to be identified and supported, especially when experiencing diabetic crises.


There is a substantial research literature on family care-giving, much of which is focused on a carer viewed as responsible for a dependent child, aged parent, or incapacitated spouse (Nolan et al. 1996, Stalker 2003, Phillips 2007). This paper describes aspects of family care-giving from a different perspective, that of adults generally viewed as responsible for their own care, but who had been admitted to hospital for emergency treatment because of their diabetes. It demonstrates the ways in which family members were involved in everyday self-care of diabetes and also in managing diabetic crises, and it highlights the importance, diversity and complexity of family support in diabetes care.

The treatment and care of diabetes have been identified as major challenges for healthcare systems worldwide (World Health Organisation 2009). In the UK, more than 2 million people have a diagnosis of diabetes, and the incidence of both Type 1 and Type 2 diabetes is rising (Department of Health 2007). In addition to the very considerable human costs of the disease, including increased risk of cardiovascular disease, kidney failure, blindness and lower limb amputations, there is a high cost to the health service in treating diabetes crises and complications (Morgan et al. 2010). In recent years, there has been an increased emphasis on the need for improved self-care and partnership working within community services, to avoid unnecessary admission to hospital for urgent care (National Health Service 2008). It is estimated that the clinical care of people with diabetes accounts for 10% of the UK National Health Service (NHS) inpatient resources (National Diabetes Support Team 2008).

The research reported here was carried out as part of a wider qualitative study to investigate pathways to urgent care experienced by people with diabetes. In addition to investigating the circumstances leading to urgent admission, we asked about wider aspects of diabetes self-care and self-management, as factors such as inappropriate diet, exercise and adherence to medication are understood to influence control of the condition.

A considerable amount of research has been carried out on diabetes self-care and factors associated with adherence to regimes recommended by health practitioners (World Health Organisation 2003, Ingadottir & Halldorsdottir 2008). Family support has been identified as one important factor. For example, Rosland and Piette (2010, p. 8) suggest that “families often create the practical, social and emotional context for self-care, making it easier or harder for patients to achieve their health and behaviour goals.” As Piette (2010) states (p. 4), “The next advance we need in the current century is a recognition that for many patients, ‘self’ management is a misnomer, since their disease care is actually shared by their family and broader social network.”

Detailed attention has been paid to specific family influences on diabetes self-care in particular communities. For example, Denham et al. (2007) researched family influences on dietary routines of people with Type 2 diabetes in rural Appalachia, and Chesla and Chun (2005) investigated the ‘accommodation’ of Type 2 diabetes in the Chinese American family, in relation to food and other aspects of self-care. Rosland et al. (2010) conducted a survey of over 400 adults with diabetes or heart failure in the US, to investigate whether families hindered self-management of their condition as much as they helped. They found that family involvement was common, and generally appreciated, although family barriers to self-management (such as encouragement to eat ‘unhealthy’ food) were also identified.

A number of other authors have explored problems associated with family influences on self-care, many of them concerned with adolescents and their parents (Dashif et al. 2008), or couples (Bailey & Kahn 1993, Trief et al. 2003, Fisher 2006). However, there has been little qualitative work on other aspects of family involvement with diabetes care, including families dealing with diabetic crises that result in hospital admission. In addition, there is little published work on the range of family members associated with specific aspects of diabetes self-care, or on support provided by family networks (see Gleeson-Kreig et al. 2002 for more details).

In this study, we learned from people admitted to hospital because of their diabetes about family involvement of different kinds – what family members actually did, day-to-day and in emergencies – and also about the range of family members providing care, illuminating the complexity of family relationships and social networks involved. Our findings are discussed in the context of current debates about family care-giving, and we suggest implications for health professionals involved with diabetes self-care.


We conducted interviews with 45 people who were admitted to hospital because of a diabetic emergency in 2009–2010.

Participants were recruited from two sites in contrasting locations in the UK chosen to maximise diversity of participant backgrounds. One site was located in a large city, with a diverse and ever-changing population particularly in terms of ethnicity. The second site comprised urban and rural districts characterised by a stable and less ethnically diverse population. The people we interviewed reflected this contrast.

People who had been admitted to hospital as a result of a diabetic crisis were identified by specialist diabetes health practitioners on hospital admission or in follow-up clinic appointments, and asked whether they would be interested in taking part in the study. Participants were given details of the research, time to consider the implications of their involvement and the opportunity to ask questions, before their informed consent was obtained. Ethical approval was granted by the relevant Research Ethics Committee for conduct in the NHS, and research governance clearance was granted by the appropriate authority for each hospital.

The range of participants interviewed was wide in terms of age, gender, diabetes type, length of time since diagnosis, medication and number of previous admissions associated with diabetic emergencies (see Table 1). We were not able to interview patients who spoke little English, except for two Bengali-speaking participants for whom an interpreter was available. We excluded under 18s, patients newly diagnosed with diabetes, those admitted with gestational diabetes or because of a cardiovascular event, and patients with significant communication difficulties. Recruitment continued until data saturation was achieved and no new themes emerged.

Table 1.   Characteristics of participants
 Number (n = 45)
 Up to 35 years 10
 36 to 59 years17
 60 years or over17
Years since diagnosis of diabetes
 1–5 years9
 6–20 years24
 21 years or more11
Reason for admission to hospital
 Hyperglycaemia or DKA21
 Foot problems13
 Other diabetes-related problems2
Diabetes type
 Type 117
 Type 227

Interviews were based on a semi-structured interview schedule. At the start of each interview, we asked participants to tell us how they came to be in hospital on this occasion, and most participants responded with a narrative account of events leading up to their admission. Later questions and associated prompts covered a variety of subjects including participants’ experiences in hospital, their diabetes history and self-care, and also family support. References to family involvement occurred in response to several questions.

Three researchers carried out the interviews. All are primarily researchers rather than health professionals, with a range of experience in qualitative research methods. Interviewers were not known to participants prior to data collection. Two interviewers each carried out interviews within a single site, and the lead researcher also interviewed people at one site as well as providing guidance and supervision across the study to ensure consistency of approach. Although the individual backgrounds and interests of interviewers differed, with two particularly interested in caregiving, similar accounts of family involvement in diabetes care were found across the full set of interviews from both sites.

Most interviews were carried out on wards or in outpatient clinics, although some took place in people’s homes, to suit participants’ convenience. Interviews varied in length depending on how much participants wanted to tell us, but most took between 30 minutes and an hour. The length and detail of participants’ accounts did not vary consistently according to the particular research setting. Family members were present in a small number of interviews; analysis of these did not indicate substantially different findings. We did not interview family members about their own perspectives on family care.

Interviews were recorded, transcribed and then coded, using qualitative data analysis software (NVivo 8, QSR International Pty Ltd, Doncasters Victoria, Australia) to help organise the data. We analysed the content of the interviews thematically, with particular reference to framework analysis methods (Ritchie & Spencer 1994). A coding scheme was agreed by the team, based on topics that emerged during the initial coding stage as well as questions asked during the interviews. We produced charts and maps to explore the data, and identified major themes, after referring back to complete interviews to ensure that fragmentation of data did not skew our interpretation at the expense of narrative and context. Results from the two sites were compared and agreement was reached about themes common to both sites. Differences in interpretation of the data were discussed and resolved in meetings of the research team, and a central motif or synthesis emerged (‘somebody there to watch over you’) as characterising the findings on families and diabetes care.

The rigour of the study was addressed following Sandelowski’s (1986) principles of auditability (a clear decision trail) and credibility (checking for representativeness of the data as a whole, ensuring typical and atypical cases).


While investigating pathways to urgent care for people with diabetes in the context of their management of diabetic conditions, we learned about family involvement with diabetes care in some detail. Although participants’ situations were diverse, a number of themes emerged, which applied broadly across both research sites.

We identified three distinct themes related to the ways in which people with diabetes described families as being involved in their care. First, family members kept a measure of vigilance, looking out for the person and being ready to take action if it became necessary. Second, they influenced everyday practices associated with routines, diet and exercise, all critically important in self-care for people with diabetes, as well as providing advice and emotional support. Third, some family members were directly involved with diabetes-specific care, such as taking blood glucose readings and giving insulin injections.

In addition, we found that the range of family members involved with diabetes care was diverse, with both co-resident and non-resident family members providing support. Participants who did not have families able to look out for them appeared to be especially vulnerable in times of crisis.

Looking out for the person with diabetes

…my mum’s scared to go on holiday... she’s scared to leave me... she gets just worried about me all the time... (Site B, Patient 10)

Constant vigilance was a feature of family care for people with diabetes. We learned about many occasions when interview participants required urgent help from family members. More than half the people we interviewed told us that a relative or friend helped to manage a crisis that led to the person’s admission to hospital for emergency care, by contacting a health professional, calling an ambulance, or taking the person to hospital. For example:

Well it was a Saturday evening and I wasn’t feeling too well. I didn’t even know that people had knocked on the door, I was just in oblivion, and my daughter came in and she said mum, I’m taking you into hospital. (Site B, Patient 12)

Monday I forgot to take my insulin … and my partner come home (sic) and she found me in a collapsed position… So she took my blood, my blood level … And it was 36.6, so that’s sky high… So she phoned up the ambulance and I’ve come in here, that’s all I can remember. (Site B, Patient 13)

Two accounts described sisters ‘taking over’ when the person with diabetes became ill and confused. Another participant told us about a relative in another town arranging for police to break in, when the relative had been talking with him on the phone and the participant became incoherent and then silent. There were also examples of young children ringing for help when a parent had collapsed.

Sometimes a hospital admission could be averted by a family member taking prompt action:

…when I’ve had a hypo, he would bring me out of it… he knows what to do, which is good, because with the hypo, I sometimes don’t know what’s happened… (Site A, Patient 3)

Family members needed to be constantly on stand-by, ready to intervene when necessary. This was not easy to manage:

I receive a lot of support from my son… [but] his place of work… they have to lock their mobiles in a locker… so during working hours I can’t get him… (Site A, Patient 18)

Reliance on family members assumes their knowledge of what to do in a crisis, and support available to them if the situation is beyond their control. In practice, family members generally contacted emergency services, particularly ‘out of hours’, sometimes after seeking advice from the family doctor. For example:

So my husband phoned for the doctor, but couldn’t get them to understand, so I just told my husband to phone the ambulance. (Site A, Patient 3)

It was not only at times of crisis that relatives looked out for the person with diabetes:

….Because I had a sister that died from diabetes, uncontrolled… and she neglected herself completely, whereas my daughter, she makes sure that I don’t… So now my daughter, all the time, mum have you done this, have you done that? She’s making sure that I do it. (Site B, Patient 12)

This interviewee appreciated her daughter’s vigilance, but in a few cases, there was more ambivalence. For example:

My father… he’s constantly telling me to check my levels …Sometimes, it can be annoying, some days I just wouldn’t want to be listening to it... (Site A, Patient 4)

Diabetes care and everyday life

Family life influenced ways in which participants managed their diabetes, either positively or negatively. There were many examples of family members helping each other out with advice, emotional support, transport and everyday care. The person with diabetes might be a provider of care as well as a recipient. For example, the participant quoted above prepared meals:

…so by cooking for her son, I eat as well. (Site B, Patient 12)

Regular, well-balanced meals are recommended for people with diabetes, and family members helped with this. For example:

….my mother would cook more diabetic friendly meals, if you like… it wouldn’t be… loads of cakes out, and things like that. (Site B, Patient 16)

However, in other cases, family eating patterns were less helpful:

…you know, we eat rich food, we like our food and a drop of wine… (Site B, Patient 13)

I have two young boys… so it’s easier just to grab a bag of crisps with them… (Site A, Patient 2)

Exercise and other elements of self-care were thought by some participants to be easier when needs were shared:

Well, after I was diagnosed diabetic, my older sister and my father were diagnosed as well, so it’s easy enough in the house food wise, and we more or less do all the same stuff, me and her would walk the dog and that together … (Site A, Patient 1)

…if I feel down, then I’ve got someone to speak to and [we] remind each other that we’ve got to take our insulin and you know, if you forget… the other one’s there to remind you. (Site B, Patient 6)

Some people talked of their relatives providing an enormous amount of care on a daily basis. Others were generally quite independent, but glad of help at specific times:

… at the moment, I’m just finding it very difficult just even getting around. I’m lucky that I’ve got my father, who helps out. I mean, if it was someone on their own … what would they do? (Site B, Patient 17)

Diabetes-specific help

Some family care-giving for people with diabetes was specific to that condition, for example, taking blood glucose readings and administering insulin injections.

In one case, several members of an extended family knew what to do in a crisis:

… it’s just like second nature to them. You know, I’ve got my box of medicines, and everything, and they go, you’ve got to take this… they know if my sugar’s too low I’ve got to have sugar, if it’s too high I’ve got to have insulin... (Site B, Patient 9)

The knowledge, skill, and commitment required could be considerable:

I’ve got a 12-year-old daughter and she knows how to do it… she takes my blood and you know, she does my needle up for me and injects me … She usually gets my tablets out in a morning, loads my needle up before she goes to school … And she says ‘Dad, don’t forget to take your 12 o’clock’ but sometimes I do forget. You know, I just nod off and forget to take it. Then when she comes home she takes my blood and it’s hyper high or low. Well recently it’s been a bit low, under 4, so straightaway she goes to the fridge and gets me out a Lucozade… (Site B, Patient 13)

Poor eyesight is often associated with diabetes, and several participants needed help with blood glucose readings and setting insulin levels. Another complication may be foot infection which, if not quickly identified and treated, may require urgent hospital treatment, in some cases amputation. This particular risk of diabetes was not often mentioned in interviews, although one participant said that his wife regularly checked his feet. Interviews with patients admitted to hospital because of diabetic foot infections requiring urgent care are the subject of a separate paper.

Several interviewees said that they relied on information about diabetes and its treatment from family members. One explained how he fulfilled an ‘expert’ role for his newly diagnosed brother:

… I gave him a lot of paraphernalia that I had, because he’s in a different health authority… about eating properly and what is diabetes … And I advised him, I said look, when you go to see them … you need to make sure you take your wife with you, so all the training sessions you go on, that you take her with you… (Site B, Patient 16)

Diversity of family networks of support

Of the 45 people with diabetes we interviewed, 38 told us about family involvement with their care. Young adults spoke of care from their parents, older people relied on their children or grandchildren and others mentioned spouses, siblings, more distant relatives, or friends who lived in the same house or acted ‘like a brother to me’. (See Table 2)

Table 2.   Family and friends cited as providing care/crisis support
 Number of interviews (n = 45)
Daughter/daughter-in-law16 (includes 3 children)
Partner10 (7 female, 3 male)
Friends/neighbours/home help7
Grandson and his partner1

Although there were examples of dyadic relationships where one person acted as ‘carer’ for another, in many cases, family networks were wider and more diverse. One participant said that she relied on expertise from her daughter who lived on another continent, and there were several other instances of caring at a distance.

Several ‘young carers’ were mentioned, and often, several members of a family – even several generations – were involved in some way. For example:

The routine in the house is, if I don’t phone my mum by 8 am, she phones me to see that everything is alright… but, on one particular morning, I had phoned my mum and told her I was up, I was grand … and I was going about, getting my daughter organised for school, and the next thing I just blacked out and collapsed……my daughter, who was then 3 years of age, she phoned 999 and asked for an ambulance... (Site A, Patient 17)

Although female relatives were mentioned most often by both men and women, there were also many references to male relatives providing support to male and female people with diabetes. Often, the type of help provided reflected traditional gender roles, but there were exceptions, for example, where a lone father provided meals and care for his adult son.

People we interviewed who lived alone and did not have family support described their situation very differently. One participant talked of struggling to fend for himself, when it was assumed that his wife (herself disabled) would care for him. One participant said she was the only person in the family who knew about diabetes, and she preferred to keep her problems to herself. Most people without family support were men living on their own. Sometimes friends were looking out for the person with diabetes. In other cases, neighbours found the person collapsed at home or in the street, and called an ambulance. One participant suggested that it was taken for granted that people have families looking out for them:

I think the way the medical people see it … is that you are attached to someone, and that there’s somebody there to watch over you… No, no, there’s no family, so it’s not that they wouldn’t, it’s just that there’s no-one there.. (Site B, Patient 21)


In this study, we found that family care for people with diabetes included a wide range of tasks, roles and relationships. In some cases, the role of children and young people was particularly important. In addition to being involved with everyday self-care of diabetes, family members needed to maintain constant vigilance, and they were key decision-makers when patients became very unwell. We found that people living alone were especially vulnerable.

A great deal has been published on family care-giving, although “the dominant overarching narrative of care-giving has become one of tragic heroine and needy client”, Barnes (2005, p. 177). In contrast, the person with diabetes is assumed to be responsible for self-care, under the guidance of health professionals. However, debates from recent published literature on care-giving provide insights that we found helpful in interpreting data from the current study.

Family members of people in our study were often able to intervene in diabetic crises. It was clear that, in many cases, family members were looking out for the person with diabetes, ready to take action when necessary. Constant vigilance may be a particularly stressful and demanding aspect of diabetes self-care for individuals and for families (Nolan et al. 1996, in relation to Rolland 1988, 1994). As Ingadottir and Halldorsdottir (2008, p. 607) suggest, some “people with diabetes deal constantly with a perceived threat and warnings about their vulnerability”.

Rosland et al.’s (2010) survey of adults with diabetes or heart failure covered aspects of family involvement with self-care. Many people we interviewed mentioned similar topics, including positive and negative influences on diet and exercise, assistance with testing and medication, emotional support and help with information and decision-making. We also found that a range of more complex and subtle aspects of family life affected our interviewees’ self-care. Our findings illustrate how coping with diabetes was only one component of lives that were multi-faceted, complex and demanding. While discussing their diabetes, participants in our study also talked about themselves, their families and friends, and their work. For many people, a role within the family was presented as important both to them and to others – even when diabetes severely affected what they could do. According to Nolan et al. (1996), a key aspect of care-giving in the context of chronic illness or disability is helping people get the most from their lives, rather than being concerned solely with illness. A positive sense of identity and taking part in family life on a reciprocal basis are important elements in maintaining quality of life.

The limitations of our sample need to be taken into account in interpreting the results of our research. We did not ask family members for their perspectives on family support, and this is an important topic for future research. The perspectives of family members may be quite different from those of the people we interviewed, as Grant (2003, p. 105) points out, “families make use of strategies to render invisible from their … relatives many of the things they do on their behalf”ey of adults with d. More generally, we did not seek to recruit a representative sample of people admitted to hospital because of their diabetes, and so our findings illustrate issues rather than drawing conclusions based on numerical data. We did not collect detailed data about participants’ diabetes, but it became apparent during some interviews that the health of the person interviewed had been affected by pre-existing diabetic complications, which might require unusually intensive family involvement in self-care. To explore whether our findings apply more generally to people with diabetes or other chronic conditions, further research is required. In addition, it would be useful to investigate in greater detail how family involvement in self-care is related to factors such as age, gender and ethnicity.

In this study, we found many instances of shared diabetes care and participants also told us about support from a variety of people within the family, including co-residents and members of extended families living much farther away. Eley (2003) emphasises diversity among family care-givers, and Lloyd (2003) suggests that examining networks of care is helpful in looking beyond the stereotype of a lone carer and service user. As Sevenhuijsen (2000) suggests in relation to social policy and informal care, it is helpful to focus on interdependence rather than autonomy, as well as to engage with people’s everyday practices.

The subject of ‘young carers’ for people with diabetes requires further exploration, particularly the impact on children and young people and their relationship with their parents.

The absence of family support networks in some cases was also significant, especially in relation to suggestions that health professionals assumed that such support would be available. People living alone appeared to be especially vulnerable at times of crisis.

It is usual for policy documents and guidance on diabetes care to include brief references to family carers (NHS 2008, 2009). However, there is very little detail about the extent or importance of family involvement. Recommendations to health professionals on structured education for self-management of diabetes refer to inclusion of family members (Department of Health and Diabetes UK 2005, National Health Service 2009), but there is little attention paid to how this may be achieved, taking into account issues of practicality and of acceptability to patients and to family members.

The issue of acceptability of family involvement in diabetes care is complex, depending as it does on diverse and multi-faceted relationships (Fisher 2006, Rosland & Piette 2010).

When asked directly about support, interviewees usually cited ‘instrumental’ tasks such as cooking and collecting medication. However, readiness to act in a crisis was also a critical element of family care for someone with diabetes (in contrast to many other long-term conditions) as it could be life-saving and required quick and decisive action. As potential crises are varied and unpredictable, preparing family members to deal with them is not straightforward. Nonetheless, it is an important issue for care planners to consider. By asking what happened to result in admission to hospital, we learned that in most cases, it was a relative who took action when the person became too ill to make decisions on his or her own behalf. At this point, although other potential sources of support might be contacted, emergency services and inpatient care often appeared the only feasible option. People without family support were very much at risk.

Conclusion and implications for practice

Our findings demonstrate that family support can be a critical component of self-care for people with diabetes, in everyday management of the disease and also in times of crisis. Family support has been shown to be associated with improved patient outcomes (Rosland 2009, Rosland & Piette 2010). It is important that family members are themselves sufficiently supported to fulfil their role and that those people who do not have relatives to look out for them are identified and supported, especially in relation to crises.

With increasing numbers of people diagnosed with diabetes and co-morbidities, and changing demographics in which there are more frail, elderly people in society, the need for family help is likely to grow. As Piette (2010, p. 4) suggests, “large healthcare systems and policy-makers are beginning to see the wisdom of supporting informal caregivers’ involvement in chronic illness care”. It can be challenging for health professionals to involve family members in education and other kinds of interventions, or to allow for the diversity, complexity and subtlety of family life. However, unless these difficulties are overcome, self-management of diabetes will remain less than optimal, with associated social and personal costs of diabetes-related crises.

More in-depth research is needed on how diabetes care is managed by both co-resident and more distant families, and what support they require. In addition, the situation of people with diabetes who live alone requires further investigation.

In summary, we suggest that it is essential to take into account the complexities of family support for adults with diabetes, if the challenges of improving diabetes care, and averting dangerous and costly diabetic crises, are to be met.


This paper presents independent research commissioned by the National Institute for Health Research (NIHR) under its Research for Patient Benefit (RfPB) Programme (Grant Reference Number PB-PG-0407-12243). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. Thanks to participants and to the health professionals who assisted with recruitment. Thanks also to colleagues, particularly Dr Adrian McCann who conducted the interviews in one of the two sites in which the research took place. NVivo Software (Copyright QSR International Pty Ltd, Version 8, 1999–2009) was used to assist the management of data.