What does quality of life mean to older adults? A thematic synthesis

Background Consideration of older adults’ quality of life (QoL) is becoming increasingly important in the evaluation, quality improvement and allocation of health and social care services. While numerous definitions and theories of QoL have been proposed, an overall synthesis of the perspective of older adults themselves is lacking. Methods Qualitative studies were identified in PubMed, Ebsco/Psycinfo and Ebsco/CINAHL, through a search on 28 November 2018. Articles needed to meet all of the following criteria: (i) focus on perceptions of QoL, (ii) older adults living at home as main participants, (iii) use of qualitative methodology, (iv) conducted in a Western country and (v) published in English (vi) not focused on specific patient groups. A thematic synthesis was conducted of the selected studies, using the complete ‘findings/results’ sections from the papers. Results We included 48 qualitative studies representing the views of more than 3,400 older adults living at home in 11 Western countries. The QoL aspects identified in the synthesis were categorized into nine QoL domains: autonomy, role and activity, health perception, relationships, attitude and adaptation, emotional comfort, spirituality, home and neighbourhood, and financial security. The results showed that although different domains can be distinguished, these are also strongly connected. Conclusion QoL can be expressed in a number of domains and related subthemes that are important for older adults living at home. The findings further support that the concept of QoL should be seen as a dynamic web of intertwined domains.


Introduction
The need for care services for older adults living at home will increase in the coming years because of the ageing population and deinstitutionalisation [1][2][3]. More and more older adults will age in their own home and the majority of them has one or more chronic disorders [4,5]. For many of these disorders, cure is not an option. However, care is required to manage these disorders and to provide assistance with daily tasks in order to enable older adults to age at home. Care and support are provided by informal carers as well as various formal care services like rehabilitation, nursing care at home, day care, mental health and general practice care [6,7].
At the same time, care providers are confronted with care reforms, budget cuts and increasing regulations for national and local commissioning and audit procedures [8]. As a result, the accountability of care services becomes increasingly important and there is a need to establish the value of such services. One way to determine this value is by assessing the outcomes achieved at the client level [9]. Maintenance of QoL is one the most important outcomes of care services for older adults. Several international action plans on ageing endorse the importance of QoL [10][11][12][13] and international interest in the measurement of QoL of older adults is growing [14][15][16][17].
It is, however, not evident how QoL should be defined or how it should be assessed. The debate about the definition of QoL is conducted among researchers from various disciplines and overlaps with explorations of the concepts successful ageing, subjective well-being, life satisfaction and happiness [18]. A taxonomy of the conceptual development of QoL shows that a large set of QoL frameworks exist, and from this set it was concluded that "QoL is inherently a dynamic, multi-level and complex concept, reflecting objective, subjective, macro-societal, and micro-individual, positive and negative influences which interact together" ( [19], p.46).
While numerous definitions and theories of QoL have been proposed, a systematic overview of the opinion of older adults themselves is missing. Knowing what older adults themselves find important in life, is necessary to align the goals of care services to their expectations. Also, knowing what quality of life is from the perspective of older adults themselves is necessary for assessing the content validity of existing QoL measures. Qualitative studies can help researchers and decision makers to understand what QoL means to older adults, and a considerable number of such studies has been done. However, these studies are largely neglected, which may be due to the fact that each of these studies has been done in a particular setting with a particular study population and particular point of view [20,21].
In order to assemble a more comprehensive picture of QoL, the findings from multiple qualitative studies can be combined in a synthesis that provides a range and depth of meanings, experiences, and perspectives of participants across contexts [22]. The larger scope and rigor of a synthesis compared to an individual study also means that there is a greater potential to influence policy and inform practice [20,[23][24][25][26]. A recent systematic overview, including all relevant qualitative studies on QoL from the perspective of older adults, is lacking. Earlier reviews did not (exclusively) focus on qualitative research [19], or were not systematic and included only a limited number of studies [18].
A method that is considered appropriate to aggregate findings from a large number of qualitative studies is the thematic synthesis approach [22,24,[26][27][28]. This approach uses thematic analysis techniques to bring together and integrate the findings of multiple qualitative studies by identifying the main, recurrent or most important themes [26]. The aim of the current review is to synthesize the findings of qualitative studies that explored what QoL means to older adults living at home.

Search strategy
A review protocol was developed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA)-statement (www.prisma-statement.org). PubMed, Ebsco/PsycInfo and Ebsco/CINAHL were searched from inception up to 28 November 2018 (by KMvL and JCFK). The following terms were used (including synonyms and closely related words) as index terms or free-text words: 'aged' and 'quality of life' or 'satisfaction' and 'narration' or 'understanding' and 'qualitative research' or 'focus groups'. The full search strategies for all databases can be found in the Supplementary Information (S1 File). Duplicate articles were excluded. All languages were accepted.

Selection criteria
For inclusion in the review, articles needed to meet all of the following criteria: (1) focus on perceptions of QoL, (2) include older adults living at home as the main participants, (3) use of qualitative methodology, (4) conducted in a Western country, (5) published in English and (6) not focussed on specific patient groups. In order to avoid omitting research of potential value to the synthesis, qualitative methodology (criterion 3) was broadly operationalized as the use of open questions and a description of the findings in words rather than numbers [29]. Criterion 4 and 5 were used in order to keep the number of papers manageable. There was no restriction to publication year. Exclusion criteria were: participants from specific patient populations such as diabetes or cancer patients (as these studies tend to focus on disease-specific aspects of QoL), and participants living in residential facilities such as care homes, nursing homes or retirement homes.

Data extraction
All papers were read several times before and during analysis by two review authors (KMvL and MSvL). The following study characteristics were extracted by KMvL: information about the sample and sampling procedure, information about the data collection method and the methodological orientation or data analysis approach, information about the focus of the study and questions asked to participants, information about the theoretical frameworks that guided the development of questions or interpretation of findings, and the main conclusions from the authors.
The complete 'results' or 'findings' sections from the studies were seen as data for this review and entered verbatim into a Microsoft Excel database by KMvL or MSvL, simultaneously dividing the text into smaller but meaningful fragments. These fragments thus included quotes from participants as well as text written by the primary studies' authors such as interpretations and clarifications of quotes, descriptions of participants' responses and descriptions of the context in which responses were given. The labels, categories or themes under which the authors described the data were noted next to the specific fragments. Data from discussion sections (including more theoretical interpretations) was not extracted.

Thematic analysis
We adopted the three stages of the analysis in thematic synthesis as described by Thomas & Harden [26]: free coding of the findings of primary studies; organisation of these free codes into related areas to construct descriptive themes; and development of analytical themes. The stages overlapped to some degree.
For the first stage of analysis, free coding, two authors (KMvL and MSvL) independently coded each extracted fragment according to its meaning and content. Within the fragments we identified QoL aspects (elements of older adults' life that affected their QoL), and used these as codes to label the fragments. Each fragment was coded using at least one label, but more than one was possible as fragments could contain descriptions of multiple QoL aspects. The labels we used were primarily based on (description of) quotes of the participants rather than on labels used by the authors in the primary studies, since the categorization of similar quotes could be different, given that these authors had various disciplinary backgrounds or theoretical frameworks. After coding the first three papers, and subsequently after every two papers, KMvL and MSvL met to compare their codes, making sure that interpretations of codes were aligned. During the process of analysis, the coding scheme was discussed extensively and continuously, and adjusted and complemented if necessary. After 11 papers the definitions of codes were clear to both authors and only few new codes came up, so the remaining papers were divided between KMvL and MSvL to code them. Occasional new codes or updates of definitions were discussed, and applied to previously coded fragments.
In the second stage of analysis the free codes of all identified QoL aspects were combined or grouped in a total of 60 descriptive themes. In the third stage these were further categorised by KMvL and MSvL into 16 analytical themes after looking for similarities and differences between the descriptive themes and the fragments belonging to each descriptive theme. Each analytical theme was described by several subthemes. One analytical theme for example was 'life philosophy', which included fragments about 'enjoying small things', 'staying positive', 'being interested' etc. A draft summary of the findings across the studies was written for each of the 16 analytical themes by KMvL or MSvL, focusing primarily on the quotes of older adults in the fragments belonging to that theme. We subsequently reread the summaries and discussed the meaning of each theme and its relations with other themes. Disagreement or uncertainties were discussed and interpretations of the themes were validated with four of the other authors (JEB, RWJGO, FAvN, GAMW). During this stage, we identified similarities and strong connections between some of the analytical themes and decided to combine some of them. This final step resulted in 9 QoL domains described by 38 subthemes that cover all the QoL aspects we identified in the extracted fragements.
Finally, we examined to which extent the 9 QoL domains were covered in the data. We determined the coverage of a domain in each paper by looking at the total length of fragments that we categorized under that particular domain. Four 'coverage' categories were used: 'not mentioned', 'briefly mentioned' (in one to three sentences), 'discussed' (in a larger paragraph), and 'discussed extensively' (in more than one paragraph). We cross tabulated the papers and the domains showing the coverage in the cells, in order to get an overview of how consistently the QoL domains were mentioned across settings and subgroups of older adults.

Results
The searches initially resulted in a total of 15,758 references, following elimination of duplicates this number was reduced to 12,257 references. After screening of abstracts, 12,139 papers were excluded because they did not meet the inclusion criteria. From the remaining 118 papers, full texts were retrieved and eligibility was determined by two authors. One paper was added via reference checking. Eventually it was agreed that 48 papers met our inclusion criteria. The details of the selection process are shown in

Reviewed studies
Characteristics and conclusions of the 48 included studies are displayed in Table 1. The studies together included more than 3,400 older adults, who were living in 11 different countries. Most studies were conducted in the US (13 studies), Sweden (8 studies), the UK (7 studies) or Canada (7 studies). The samples were very diverse, including among others older widowed women living in rural and remote areas in Australia [31], foster grandparents from Missouri (US) [32], older adults receiving palliative care in Sweden [33] and older men residing in London (UK) [34]. Participants were most frequently recruited via care organisations and senior centres or via their participation in epidemiological cohort studies. In almost all studies either half of the participants or a majority of the participants was female. Reported mean ages ranged from 71 to 91 years. The most commonly reported reasons for excluding participants were minor or major cognitive impairment, not being able to communicate in a specific language, or less specific reasons for not being able to participate or give informed consent.
Nine studies used focus groups or group interviews for data collection, five studies were based on open questions in a survey, and the remaining studies used individual interviews. The method of analysis consisted of content analysis or thematic coding, grounded theory analysis, and phenomenological or hermeneutical approaches.

Quality of life domains
We categorized the QoL aspects included in the extracted data into nine QoL domains: 'Health perception', 'Autonomy', 'Role and activity', 'Relationships', 'Attitude and adaptation', 'Emotional comfort', 'Spirituality', 'Home and neighbourhood', and 'Financial security'. An overview of the domains and subthemes is shown in Table 2. Each domain is further described and illustrated with quotes in the section below.
Health perception: Feeling healthy and not limited by your physical condition. Health is mentioned as a necessary and sometimes even paramount element of QoL, eg: "Above all, being healthy is the most important thing to have quality of life, the rest comes as an extra" ( [37], p.76). The perception of health was partly determined by the extent to which older adults felt fit and active or suffered from physical, mental and cognitive disorders. Troublesome symptoms, functional limitations and side effects from medication (such as poor balance, poor memory, pain, vision loss and fatigue) significantly decreased their QoL. "Blindness is the one thing and then I was diagnosed with diabetes and then I had a heart operation. I'm still walking around and I still enjoy playing music so in that way I'm blessed but in other ways I'm not . . ." ( [77], p.944).
The perception of health is also determined by the point of reference used; older adults compare their health for instance often with that of others their age. The experience of health was therefore described as a relative phenomenon [36]: it is experienced and evaluated according to what one finds reasonable to expect, given one's age, history, medical condition, and social situation. For example, some older adults find declining ability frustrating "I cannot do what I did ten years ago, and I get very angry . . . I get disgusted with myself" ([42], p.934), while others were more accepting, saying that other people are worse off, or that pain, fatigue and illness were to be expected in old age. This explains why older adults may still perceive their health as 'good' despite chronic diseases, illnesses and frailty [59,60,67]. Some older adults are committed to improve their QoL by trying to influence their health with a positive attitude and an active lifestyle.
Health was described as important because it is the basis for many other QoL aspects: "What mostly controls it is your personal health. That determines what you can do." ([72], p.55). Good health appeared to facilitate the ability to carry out meaningful activities, to take care of yourself, to perform household tasks, to get out, communicate and participate. Murphy et al. [70] noticed that especially as participant's physical functioning declined older adults started to redefine health in terms of abilities rather than absence of illness, e.g. "Health is to a great extent being able to look after yourself" ( [49], p.292). Only when older adults were severely restricted in their 'going and doing' [42], they experienced poor health. Such experiences were connected with negative emotions of sadness, anxiety and sorrow [53]. Autonomy: Being able to manage on your own, retaining dignity and not feeling like a burden. Many older adults mention the desire to stay independent as long as possible: "If I recover from this and can take care of myself, then I shall be satisfied. It's important not to have to trouble others, and to be able to carry out the duties I can" ( [35], p.1116).
Participants say that being independent enables them to experience a sense of freedom and to enjoy life, by being able to socialise, to go outdoors and to do what you want. As Bryant et al. [42] noticed, respondents spoke with pride or pleasure of their independence, saying for example 'we can take care of ourselves' and by giving examples of everything they still do. Several participants indicated that the worst imaginable situations would be to become completely dependent on others or to end up in a nursing home.
Carrying out activities and daily routines independently contributes to a sense of control. For example, doing household task without help implies that older adults can decide Thematic synthesis of Quality of Life When you are used to manage on your own without having to take into account another person's time and availability to help, it can be difficult to accept and acknowledge the need for help. Older adults do not want to be a burden to others. Although they value they value the assistance and support they receive, they often indicate that they are afraid to be(come) a burden to others. Some therefore withdraw from contact with friends and family "You don't want to spoil theirs [enjoyment] and be a nuisance to them" ( [53], p.283) or even rather prefer the thought of passing away [35,42].
Authors of the included studies remarked that feelings of being a burden, guilt and embarrassment due to dependency affect older adults' self-image. This may be the reason why, rather than acknowledging to be dependent, many older adults characterised themselves as independent by emphasizing the things they can still do [77,78]. Focusing on remaining abilities, personal care and appearance helps to retain dignity, and when care is needed: being helped and treated with respect for one's personal identity, wishes and values. From et al. [53] and Milte et al. [68] noticed that especially partners and relatives can help the older person to preserve an experience of dignity as they are well aware of their habits and are sensitive to their wishes:  Respondents mentioned a broad range of valuable activities that brought joy or meaning to their life, including hobbies, mental and physical activities or more socially oriented activities, such as going out, travelling and participation in group activities. Indicative is the following quote: "I love the theatre, I love the cinema . . . If it's just to go off for a day somewhere and have a meal in a pub . . . And I think that's very essential, it's just the simple pleasures of life really" ( [57], p.1897).
Older adults indicate that activities such as reading the newspaper, learning new things and keeping in touch with family and friends also help to stay connected with the world around them. They value feelings of belonging, participation and having a role in society, although they sometimes feel excluded: "It's unfortunate that a lot of your family or other people feel that, once you're old, you don't know anything anymore and you're just kind of in the way." ([62], p.574) In order to feel valued and to gain a sense of purpose, older adults mentioned that they find it important to continue helping others, for example by volunteering, babysitting or caring for sick relatives. By doing so they feel they contribute to society and fulfil a role: "I think the quality of life is being involved and having a part to play. I think if you lose your role in life then you start getting depressed, I think it is very important to be needed for whatever reason, and . . . Thus, older adults' self-worth is closely related to the activities they are able to do and achievements and roles in current and earlier life. The effect of limitations on one's self-image is for example described as follows: "Yes, I think it's meaningless when I obviously can't be active in any way. . . You don't feel like a real human being in some ways, more like a vegetable or something" ( [35], p.1115). Respondents mention that it is important to have a positive picture of yourself, but that their self-worth is sometimes threatened by the feeling of being perceived as a nuisance in society.
Relationships: Having close relationships which makes you feel supported and enable you to mean something for others. Social contacts are seen as essential for QoL by older adults, they help to avoid loneliness. Especially close relationships are valued, as the quality of contact is most important: "Togetherness is understanding and trusting someone else. Being able to talk to someone about things you wouldn't speak to others about and them doing the same." ([53], p.282).
Many quotes show the importance of relations with family members; to provide help, support and love, and give joy and meaning in old age; e.g.: "How can I put this, the older you get, the more you focus on your own family and your children. [. . .]. Yes, you're there for each other." ([74], p.269). Hörder et al [60] remarked that feeling loved by close relatives and caring about them can keep older adults from worrying and distract their attention. Especially partners and (grand)children are a source of joy and support, although the bond may also bring worries, conflict and sorrow. Fear of violence and abuse were only marginally reported [37,54,76].
Experiencing support (both practically and mentally) in personal relationships was a recurring theme, and older adults talk about how they appreciate being looked after by people that care about you, having someone you can call if you have any problems, and being encouraged in times of setbacks. Some older adults who lacked close relationships expressed the wish for support: "I'd just want friendship, some compassion, understanding, empathy." ([42], p.936). A bond with pets can provide support as well [40,64].
Besides providing support, close relationships can be a source of love, affection and appreciation: "Being in the company of my dear ones gives me a sense of belonging. It's nice to know that someone cares, and that they in turn can count on me when they need help or someone to talk to [. . .] Above all, I want to love and be loved." ([36], p.43). Sexuality and intimacy were only briefly mentioned in the studies by Kalfoss [61] and Bryant et al. [42].
As shown in above quote, older adults want reciprocal relationships and mean something for others as well. Keeping relationships equitable may be especially important to maintain self-worth and diminish feelings of being a burden.
Attitude and adaptation: Looking on the bright side of life. Older adults strongly believe that a positive attitude helps to have a good QoL. Older adults who adopted such a life philosophy described their motto as: stay positive, enjoy life, be happy with small things, make the best of life, and maintain your humour, optimism and curiosity. They emphasize one should not feel sorry for oneself, complain all the time or just sit down and do nothing.
In many ways, ageing relates to 'living in a new reality' [77]. In order to maintain a feeling of wellbeing, older adults state that it is essential to accept, adapt to and cope with life changes inherent to ageing, such as retirement, slowing of pace, declining health status (of partner) and loss of loved ones. Inability to adapt may result in feelings of sadness and despair. Some older adults even equate QoL with the ability to adapt: "20 or 30 years ago, I might have listed some of the usual terms associated with the expression "good quality of life": good health, an adequate income, a reliable social network, good friends, and an enjoyable and encouraging job. Today, at the age of 86, my definition is different. With a growing absence of those pleasures, I now see a good "quality of life" as the ability to adapt to an increasingly difficult situation without letting problems interfere with gratitude and joyfulness; to experience fine literature and music, and as far as one can manage, to be useful and encouraging towards other people. In other words, a good "quality of life" is the ability to endure no matter what." ([36], p.43).
Older adults used several ways to adapt to difficult situations, including deliberate and unconscious processes. Acceptance is the strategy most often mentioned. Respondents recommend to make the best out of each situation and to take life as it comes by accepting what happens to them: "You never do get over it but if you can accept it you can start taking the steps, you've got to . . .life's got to go on, hasn't it?" ([55], p.683).
Some more unconscious coping processes that were identified include minimizing and denying limitations, laughing about difficulties, emphasizing abilities and skills that one had in the past, and adjusting one's standard of what is acceptable or important. A woman, who used to care a lot about looking representable and wearing panty hoses, now says, "But it was so much trouble. [ ([59], p.787). Older adults put their situation into perspective by comparing their situation to others, or to what is normal or expected at their age. This could entail either 'upward comparisons' (e.g. inspired by the strength, will and optimism of others) or 'downward comparisons' (e.g. feeling grateful when considering your health better than that of your peers).
Older adults also adapt and maintain their QoL by changing their behaviours and habits. Some respondents, for example explain that they decreased their level and duration of activities in order to prevent fatigue or deterioration of their health: "You have to take a rest and then you can do it again." ([39], p.205). They also said to avoid risks and to be more cautious or attentive when carrying out tasks, and to seek more assistance from other persons, technology, aids and equipment. Some even anticipated to future limitations, such as moving to live near children or investing in a good physical or cognitive condition.
Emotional comfort: Feeling at peace. Older adults indicated that they wish to feel calm, content, free from worries, in harmony with life and to have peace of mind. However, for some, these feelings are hampered by stress and worries about loved ones, low income, health and independence: "The worries that drag around with you is like a weight you put in the trunk of your car that slows you down." ([64], p.e95). Some older adults express severe levels of stress, and report to feel frightened, fear the future or have anxiety. This severely impacts their QoL.
Quite some respondents mentioned (often in relation with their positive attitude) to be happy, enthusiastic, embrace life, and appreciate or enjoy small things. Others report to feel sad, downcast, hopeless or even depressed. Loneliness, usually as a result of bereavement or moving into a new community, has a strong negative impact on older adults' QoL. Borglin et al. ([39], p.211) pointed out that the loss of a partner implies immense sorrow and a feeling of having lost an important part of oneself: "The loss is indescribable, when you have been together for so many years, but that is why I say that I have lived my life, for the fact is that I'm not living any more, I only exist." Losing connections with friends and family members is difficult as well: "My best friends are all gone. . . and I feel that is one of the worst parts of getting old and surviving: you have to bear the pain of seeing them go." ([42], p.935).
Lastly, the reminiscence of negative and positive past experiences can influence current feelings. Some memories are painful, e.g.: "These memories, they never go away. No matter what pills you take, it doesn't take away the burden that I've been carrying right up till today" ( [72], p.53), whereas others bring joy, or feelings of accomplishment and gratefulness.
Spirituality: Feeling attached to and experiencing faith and self-development from beliefs, rituals and inner reflection. Being religious or spiritual can support older adults in accepting disability or psychological distress, in coping with changes and being satisfied with life: "It helps me to cope with happenings, incidents and so on, which would seem to be absolute tragedies to other people" ([57], p.1895), and "I just am so thankful for what the good Lord has given me." ([42], p.937). Religion and spiritual beliefs shape their philosophy of life and can also relieve thoughts about death "I am not afraid of dying [. . .] I have had a good life and I look confidently forward to an even better after life." ([39], p.205). Some older adults rely on their God and turn to him to help overcome problems. Faith gives them comfort, peace of mind and a sense of meaning and purpose in life, and makes them look forward to each day. Volunteering and taking part in religious activities and practices, such as going to church, were described as ways to stay socially active and involved.
However, there was large variation in the value older adults attributed to religion and faith. Some respondents commented that religion was not important at all: "Well I haven't any faith in a personal God that's looking after you. If you know anything about astronomy you realise how stupid that is." ([57], p.1895).
Spirituality is often mentioned as synonymous with religion. Yet, it may also encompass non-religious beliefs and rituals, such as meditation, self-development and turning attention to the inner life: "I'm spiritually-not religious-I'm spiritually very calm with myself. I discovered things about myself in mid to later years that have been sitting there all the time." ([75], p.570). For some older adults ageing means a certain kind of spiritual growth and becoming a wiser or better person.
Home and neighbourhood: Feeling secure at home and living in a pleasant and accessible neighbourhood. The meaning of home for older adults often extends that of 'just a residence'; it brings about feelings of staying in a pleasant place with familiar and important objects and shared norms, history and values [49]. This is illustrated by a respondent: "It's also important to have a safe place to live. It means a lot to have a flat that is adapted to your needs." ([36], p.46). A home provides shelter, a place where one can feel safe and comfortable, and it is a private domain in which you are in control and can maintain your daily routines: "my home is my castle" ([39], p.212).
Many older adults strive to keep living at home or in the community for as long as possible and preferred adapting the home environment rather than moving; "Well, the first condition is to stay fit enough to be able to live on your own. And to live at home; I'd much rather live here at home than in some institution" ( [72], p.54).
When becoming dependent on other people or services, the feeling of safety and privacy at home can be at stake. Older adults addressed how home adaptations and alarm systems increase their feelings of safety. Some were concerned about safety from crime in and around their home: "After dark, we never go out. And we never answer the door unless we're expecting somebody; otherwise, our door remains closed" ( [74], p.271).
Having friendly neighbours and experiencing a sense of familiarity in your neighbourhood can enable a sense of security: "I suppose it's security, ain't it-you've been in the area for a long time, you know where everything is, you know everybody locally-you feel more secure" ([57], p.1894). Neighbours and social networks can provide help in everyday life and social contact. Furthermore, according to older adults QoL is enhanced by a pleasant environment which provides the opportunity to enjoy nature and do interesting activities.
However, some environmental barriers were mentioned that restricts older people from going out. These are for example poor weather and obstacles such as 'poorly maintained sidewalks, and street/traffic designs poorly adapted to older adults' [76]. Accessibility of the environment is a key to participation in community life [70]. According to Bowling et al. [40], lack of easy access to reliable, cheap and convenient public transport was mentioned as inhibiting social contacts and activities. Access to community services such as post offices and libraries was mentioned to be important as well.
Financial security: Not feeling restricted by your financial situation. Good financial circumstances make life easier. Respondents mentioned their ability to meet basic needs, for example: "It's having sufficient money. . . to do what you require, run your car, say, and pay your bills, and have the odd holiday. . . sufficient money not to have to worry about money" ([41], p.838). Income affects people's capacity to pay for services such as home-help, physiotherapy and meals, or transport. It seemed that especially respondents who experienced financial difficulties mentioned finances as important for QoL.
Adequate finances are not only important to afford basic essentials and household bills, they also provide freedom: the possibility to afford outings, activities and extras. Older people can feel restricted by a lack of financial resources which hinders them to enjoy themselves and to participate in society (e.g. through holidays, going to the cinema, inviting friends for dinner, a train ride, luxuries) and to feel independent and secure (living comfortably and without worries). Respondents lacking sufficient money felt unable to enjoy life. "There are so many things I would like to do that I can't afford and that is not fun" ( [39], p.212).
Next to having sufficient money, respondents mention how material and (housing) conditions help to feel comfortable and retain independence: "I should buy myself one of those adjustable beds. And

Quality of life as a web of intertwined domains
The domains should not be interpreted as clearly demarcated and distinct units. Our findings rather imply that QoL can be understood as a dynamic web of domains, where each domain covers a swarm (or cluster) of strongly connected QoL aspects. During the analysis it became clear that the subthemes that describe the QoL domains sometimes overlap and are strongly connected with each other. Within quotations, these connections were visible by statements about how QoL aspects affect each other, for example: "If you have a good mental outlook, you are probably going to be active and you're probably going to eat reasonably well" ([63], p.s34). Several connections and points of overlap between QoL aspects are exposed in the domain descriptions above, for example, the idea that being able to mean something for other people is important for your own QoL is part of the subtheme 'helping others' (domain Role and Activity) but also of the subtheme 'reciprocality' (domain Relationships). Moreover, it is strongly connected to the subthemes 'being a burden' (domain Autonomy) and 'self-worth' (domain Role and activity).
The domains are thus intertwined; they interact with each other and sometimes even partly fuse. However, a clear hierarchical picture of QoL domains, in which some domains can be identified as purely instrumental to other domains, was not found. Health was for example said to be needed for QoL aspects covered in other domains such as 'Role and activity' and 'Relationships', but certain activities and relationships also contribute to one's 'Health perception'.
All the interactive threads between the QoL aspects make the web dynamic: if something occurs in one of the domains it influences the rest of the web. This implies that the status of a domain for an individual is affected by the status of the other domains at that time. Table 3 shows the 'coverage' of the domains in the individual studies. As such, it provides an overview of how extensive and consistently the QoL aspects that we categorised into these domains were mentioned across settings and subgroups of older adults. For example, the domains 'Autonomy', 'Role and occupation', 'Attitude and adaption', 'Health perception' and 'Relationships' were covered in almost all studies, suggesting that the QoL aspects categorised into theses domains are important across settings and groups. These QoL aspects were most detailed and extensively described in the studies and are therefore more prominent in our synthesis. QoL aspects categorised into the domains 'Spirituality' and 'Financial security' were less universal as they were less consistently covered across studies.

Main findings
The purpose of this review was to synthesize the findings of qualitative studies that explored what QoL means to older adults living at home.  Health perception Role and activity Attitude and adaptation Home and neighbourhood While we distinguished these nine domains, the quotes from older adults often fitted into more than one domain and the quotes exposed numerous connections between domains. This implies that the concept of QoL can be understood as a dynamic web of domains meaning that the domains are strongly intertwined and are affected by the status of the other domains at that time.

Reflection on findings
Although it is often claimed by researchers [53,73] that little research has been done into the perspective of older adults on QoL to motivate the relevance of their empirical study, this review proves the opposite. There is a wealth of material available covering more than 25 years of research.
All authors of the primary studies used some sort of sub-classification of the themes they identified, which confirms that QoL is perceived as a multidimensional concept. The various ways in which primary authors labelled, classified and made sense of QoL aspects mentioned by older adults can be explained by our understanding of QoL as a dynamic web of domains that strongly interact and partly overlap. However, we found that the content of the quotes from older adults themselves was very similar across the studies, showing common experiences and use of language to describe these experiences and the contribution to QoL. This means that under these diverse interpretations and conclusions there is something more universal in the meaning of QoL for older adults living at home in Western societies. Rather than using the more theoretical interpretations of QoL by the primary authors for our classification system we focused primarily on the quotes of older adults in order to grasp these universal experiences. Furthermore, our synthesis is a more comprehensive classification of QoL aspect compared to the QoL classifications in the primary studies.
In some of the included papers, domains such as 'Health perception', 'Home and neighbourhood', 'Spirituality' and 'Financial security' were characterized as instrumental to or as a resource for other QoL domains, because some older adults mention these themes merely as helping them to adjust and do the things they would like to do. In a similar vein, themes described under 'Attitude and adaptation' for example are often about strategies to improve QoL or maintain a good QoL. However, there are also fragments showing that these domains are important in itself or are influenced by aspects covered in the other domains, such as the activities that one participates in. It is therefore difficult to draw a line between domains that are more instrumental to QoL and domains that should be seen as ultimate QoL components, i.e. in cause and effect relation.
The domains 'Health perception', 'Autonomy', 'Role and activity', 'Attitude and adaption' and 'Relationships' were most extensively and universally covered across settings and participants. Although this suggest that these domains are most distinctive in characterising QoL, we think that the coverage of the domains not necessarily reflects the importance of these domains for QoL as conceived by older adults. For example, the domains 'Spirituality' and 'Financial security' were less extensively and consistently covered, but they seem very important for specific groups or individuals. Our list of domains should therefore be seen as a comprehensive framework of potentially important QoL aspects. The relative importance and meaning of the QoL domains depend on one's circumstances but also on personal preferences. This can be taken into account in a QoL measure by giving respondents the opportunity to indicate how their current situation relates to their needs in/wishes for certain domains as is for example done in the ASCOT [79,80].
The synthesis affirms that adaptation processes can result in a shift in perceived importance of domains, as well as in an altered understanding of what these domains mean to someone, also known as 'response shift' [79,80]. For example, in case of declining health and independence, older adults seem to focus their description of health and independence on their remaining abilities. Rather than a strict or technical definition in terms of physical functioning (which may result in a characterisation of themselves as 'unhealthy' or 'dependent' persons), they describe health in comparison with others of their age and in terms of ability to adapt and to carry out meaningful activities (which, in turn may shift from more difficult goals such as gardening to more easy tasks such as grooming and eating). So, both the importance an individual places on and understanding an individual has of each of the QoL domains are dynamic and change over the course of life.

Comparison with other studies
Stanley and Cheeck [18] only identified and reviewed four qualitative studies that investigated older adults' perspective on well-being, and the results of these studies were not synthesized whereas we synthesized findings of 47 studies. The activities and strategies listed in their review, however, are comparable to the domains identified in our synthesis.
In an earlier review by Brown et al. [19]), it was stated that QoL can theoretically encompass a wide ranging array of domains. The QoL components Brown et al identified from a mix of 45 surveys and qualitative studies among older adults were remarkably consistent to the themes identified in our review. Some of our subthemes like 'Home and neighbourhood' (a good home, safety, transport facilities and a pleasant environment) were mentioned in the review by Brown et al., but were allocated to other domains. Nothing was mentioned in the review by Brown et al. about 'Attitude and adaptation', while we found that older adults consider the ability to adapt to difficult circumstances essential for a good QoL.

Strengths and limitations
This study contributes to our understanding of QoL from the perception of older adults as it brings together the findings of 48 studies, covering views of more than 3,400 communitydwelling older adults from 11 Western countries. The synthesis therefore provides a more universal understanding of QoL than each study on its own. By recoding and synthesising the reported data rather than summing up a list of identified themes we aimed for a 'whole' greater than the sum of its parts.
The samples from the included studies were selected from very diverse populations, covering younger and older age groups, and active volunteers as well as vulnerable older adults. Although, we recognize that diversity of respondents/samples may influence how they characterize quality of life, our aim was to define the commonalities over these diverse respondents. Therefore, we analyzed the content of respondents' quotations, not their personal characteristics. Considering the fact that diverse populations were included, we are confident that our findings cover a wide range of views on what QoL means to older adults. Although some QoL domains prevailed more often in specific contexts, the identified domains were fairly consistent across the studies. This contributes to the credibility of the findings.
In our study and in the selection of articles, we did not use a fixed definition of QoL, but rather we employed a broader view on QoL. This is among others shown by our broad search which included search terms such as 'life satisfaction', 'successful aging' 'living well', 'wellbeing'. Our goal was to include all studies describing the perspective of older adults on quality of life, and not to engage in the theoretical discussion about existing definitions. An advantage of using such a broad approach is that the QoL domains presented here are all based on what is important to respondents themselves. A disadvantage might be that it is not clear how the domains identified in our study relate to other concepts of QoL, well-being or successful aging. Our broad search strategy is reflected in the fact that we only identified one additional paper from the reference list of one of the included studies. Thus, we are confident that we included virtually all relevant studies.
Because we found that similar QoL aspects were identified across the studies and because our coding scheme did not change much after analysing the first half of papers, we think that our QoL framework does not need to be revised considerably when new qualitative studies on this topic will be added. There may however be themes that older adults or interviewers are reluctant to address. One subtheme for example that is surrounded by an 'audible silence' [24] is intimacy and sexuality in old age. Although large parts of the texts in the papers concern relationships, hardly anything was reported about the importance of experiencing intimacy. It might be that this is a theme not considered important by older adults, but it also possible that they are reluctant to share their thoughts about this topic in interviews that focus on QoL [81].
To enhance the transparency about the origin and background of the QoL aspects identified in our synthesis we provided a table showing which QoL domains were covered in which papers (Table 3) and a table showing the 'signature features' [21] of the included studies, such as the sampling strategy, the focus of the study and the methodological orientation and theoretical frameworks used. Individual interviews as well as focus groups were included. The variety in signature features supports the conclusion that are findings are not influenced by one specific methodological orientation or theoretical framework.
However, there are limitations to condensing such a wealth of material. Sandelowski et al.'s question 'Can you sum up a poem?' ( [21], p.366) very well reflects the challenge of synthesising qualitative studies. A synthesis undeniably leads to a selection of material which in itself already is the result of a process of condensing and interpretation of primary raw data done by the authors of the studies. We therefore refer readers to the individual papers if they are interested in more detailed discussions and interpretations of the identified domains. Tables 2 and  3 can serve as a guide for identifying relevant studies. Opponents of synthesising qualitative studies remark that the loss of explanatory context is a cause for concern as qualitative findings are often specific to a particular context, time and group of participants [21,23]. We attempted to preserve some context by providing summaries and signature features of each study in Table 2.
A second caution concerns our own perspective and professional and personal backgrounds. Our classification of QoL domains as resulting from the synthesis is influenced by our own values, background and experiences, just like the various classification systems as constructed by the primary authors of the included studies. Our research is mainly health-oriented and less socially oriented, which likely shaped our interpretation of the experience of QoL by older adults living at home. Yet, the possible effect of our personal backgrounds is reduced by the variation of the authors and the thoroughness of the peer review process in all phases of the study.
Finally, we did not perform a formal quality appraisal of the included studies. Researchers vary in their opinion on the use of quality appraisal of qualitative studies in a synthesis [20,21,23,24,26,28,[82][83][84]. Apart from the diverse opinions concerning what a 'good' qualitative study is, there is no consensus about the most appropriate role of such an appraisal. For example, Atkins et al. [23] experienced that appraising studies became an exercise in judging the quality of the written report rather than the research procedure itself. Campbell et al. [24], therefore, argued that only studies with fatal flaws should be excluded. Based on a subjective overall evaluation going from our personal experience with doing and reporting qualitative studies, we did not consider any one of the included studies as fatally flawed so did not see a reason to exclude one of them from the synthesis. Furthermore, because papers with mainly short descriptions and explanations of the data offer in general few insights, in several studies it was found that poorer quality qualitative studies contribute only minimally to a synthesis, adding weight to the better studies [23,24,26]. We found in our study as well that the studies that provided the most conceptually rich descriptions and comprehensive quotations of QoL aspects (as indicated in Table 3) automatically contributed most to our synthesis. Finally, assessment of quality is much more important when themes of the included studies are used. Since we did not use the codes used by researchers of the included studies, but defined a set of themes and codes ourselves that we applied to the data extracted from the included studies, we considered it unnecessary to assess the quality of the included studies. Stanley and Cheek's [18] conclusion that the older person's subjective perspective on wellbeing is lacking in the literature, does not hold anymore given the fair amount of studies that we have identified. Our review provides a guide to researchers by providing an extensive overview of individual qualitative studies providing a complete overview of QoL domains and subthemes and references to more specific sources.

Implications for care practice, measurement of QoL and further research
Our findings show that autonomy and living at home are valued by older adults, which is in line with current ageing in place policies. Because the QoL web is dynamic and consists of intertwined domains, such policies likely affect other QoL domains and may result in negative consequences as well, for example in loneliness and safety. With more and more older adults ageing at home there will be an inevitable increased need for care services and support. Care and support for older adults living at home should be aimed at improving or maintaining QoL, foremost on the QoL domains that they value themselves. Perhaps the most important practical implication of this review is that a broad definition of QoL should be targeted, encompassing various interrelated domains. The identified domains can provide direction to care services and support, and consequences on all domains should be considered when developing personal care plans. These domains can be used in personal conversations about how older adults' QoL could be improved and to personalize their care. Because the QoL web is dynamic such plans should be updated and discussed regularly. Furthermore, although it can be discussed to what extent care services and support are responsible for all identified QoL domains in this synthesis, the domains should be considered as outcomes or important side effects in improvement processes, evaluation, monitoring or benchmarking of care services and support for older adults living at home.
The findings of our study may also be used by researchers developing new QoL measures for older adults. Currently, not all of the identified domains are sufficiently covered in existing QoL measurement instruments for older adults. Many available instruments focus on healthrelated QoL and include physical and psychological dimensions [15,16,85,86], especially the preference-based measures developed for use in economic evaluations [14,17]. And preference-based measures adopting a broader perspective on QoL do not include a health dimension [87,88]. Although these measures may be suitable for specific purposes, our findings suggest that from the point of view of older adults important domains are missing. Further research on the operationalisation and measurement of QoL domains least frequently covered (Autonomy, Role and activity, Attitude and adaptation, Spirituality, Home and neighbourhood, and Financial security) is recommended in order to increase the face validity of QoL instruments for older adults. Development of measurement instruments based on a broad definition of QoL in line with older adults' perspective may help guide care services to direct their policies at what is important for older adults.
Nonetheless, several findings from this review indicate that incorporating older adults' perspective in the measurement of QoL can be challenging. Especially the dynamic character of QoL and shifting reference points and concept definitions over time by older adults, suggest that it may be problematic to use generic, static instruments. Adaptive, flexible ways of measurement are probably more in line with the characterisation of QoL found in this review. We also recommend to carefully align the choice or development of instrument to the goals and setting of measurement.
Furthermore, the interrelatedness of domains and absence of strict boundaries between domains mean that for older adults, any classification of QoL domains may feel artificial. As Hendry & McVittie [58] put it: "people do not segment their lives into component parts".

Conclusions
We identified nine QoL domains from 48 qualitative studies among different groups of older adults. Older adults value feeling healthy and not limited by their physical condition, being able to manage on their own, retaining dignity and not feeling like a burden, spending time doing activities that bring a sense of value, joy and involvement, having close relationships which makes them feel supported and enable them to mean something for others, looking on the bright side of life, feeling at peace, feeling attached to and experiencing faith and self-development from beliefs, rituals and inner reflection, feeling secure at home and living in a pleasant and accessible neighbourhood and not feeling restricted by their financial situation. Which domains apply in a specific situation needs to be decided by service providers and care professionals themselves, preferably by turning on specific domains using flexible measurement instruments. However, it is important that service providers and care professionals realize that the QoL domains are strongly intertwined meaning that changes in one domain likely affect other QoL domains.