Research team

A team of Lived Experience Researchers, (LERs), drawing on their own experiential knowledge about living with mental health problems, and other researchers from the UKRI Loneliness and Social Isolation and Mental Health network and the National Institute for Health and Care Research Mental Health Policy Research Unit (MHPRU) planned and conducted the study. The team included clinical academics and non-clinical researchers from a range of backgrounds (including qualitative and mixed methods research, health policy, health economics, and the arts). The research team met weekly by Zoom video call [27] to plan the study and discuss progress. Most interviews were conducted by thirteen LERs involved in the study, except for eight telephone interviews conducted by MB, an experienced qualitative researcher and occupational therapist. Three LERs were employed in university research roles; others had honorary research contracts with University College London. Eleven of the LER interviewers were female and five were from minority ethnic backgrounds. The LERs received training on conducting face-to-face and online interviews and obtaining written and verbal informed consent. A weekly lived experience reflective space provided LERs with emotional support and space to discuss the research process and emotional impact, peer-facilitated by four experienced LERs.

Sampling and recruitment

Purposive sampling was used to ensure diversity regarding participants’ diagnoses, use of mental health services, and demographic characteristics, including age, gender, ethnicity, and sexuality, and whether they lived in rural or urban areas. We reviewed our sample during recruitment and implemented targeted strategies to ensure diversity. These included approaching community organisations working with Black and Minority Ethnic communities and using targeted recruitment materials.

Participants were eligible to take part if they were aged 18 years or over, had a self-reported mental health problem and lived in the UK. We recruited through three London-based community organisations (a mental health charity, a homeless charity and a community arts organisation), as well as through social media, especially Twitter, supported by the Mental Elf. Several charities and community organisations supporting people with mental health problems also agreed to disseminate an invitation to participate to their networks. Potential participants contacted the research team by email. Researchers then checked eligibility, provided a participant information sheet, answered questions about the study, and booked interviews.

Data collection

The topic guide (see S1 Appendix) was developed collaboratively by members of the Loneliness Network’s Co-Production Group to explore the nature of loneliness experiences, their relationship to mental health, and alleviating and exacerbating factors. Prompts were included for questions asked to ensure topics were fully explored. Following the onset of the COVID-19 pandemic, further questions were added regarding experiences of living with mental health problems during the pandemic (see S2 Appendix for revised topic guide). Semi-structured interviews took place between March and July 2020. Ten participants were interviewed before the introduction of COVID-19 infection control measures, (two face-to-face with LERs and 8 by phone with MB and 49 participated in online interviews with LERs following the pandemic’s onset. Informed consent (verbal or written) was obtained prior to all interviews. All interviews were audio-recorded, with verbatim transcripts produced by an external transcription company. All transcripts were then checked by the researchers and any identifying information was anonymised.

Analysis

We took a participatory approach to the analysis as a large team of researchers from varying backgrounds. We used Template Analysis [28], a form of thematic analysis [29] involving a codebook approach. Analysis involved defining and organising themes using a coding template, which was developed and refined during data analysis through an iterative approach. All interview transcripts were analysed by four Lived Experience Researchers (RRO, BC, PS and PN) and a network researcher (MB) facilitated by NVivo 11 software [30]. MB undertook a preliminary analysis of three transcripts, reading and re-reading to identify initial themes. Three LERs independently analysed one each of those transcripts and discussed points of divergence. Differences were compared with MBs initial themes identified to highlight areas requiring closer examination and to ensure the coding captured complex and nuanced data relevant to the research topic. The aim was to capture richer data, to guide further coding, and not to seek a consensus on meaning, in keeping with the approach of reflexive thematic analysis [29]. All themes formed an initial coding template. A further set of five transcripts were then analysed individually and the further initial themes were then discussed between the researchers conducting the analysis. This group reflected on the themes together to ensure important ideas in the transcript had not been overlooked, and to refine the initial coding template.

The analysis followed an iterative process where the core coding team met fortnightly with the wider research team to discuss and refine the developing list of themes in the coding template, obtaining further perspectives on the analysis. All transcripts were then coded following the final template.

Theme one: What the word “lonely” meant to participants

Loneliness was described in a variety of ways, including as an emotional or physical state, or through accounts of lack of connection, feeling of belonging, or love and support. A unifying element was that it was not just an unwanted lack of contact with people but had important psychological elements.

1.3 Lack of a sense of belonging

Some people described loneliness as related to objective isolation:

“Some people… they have no choice… they don’t have the care and support from family and friends.”

[P38, male, Black, 46–55, urban]

For others, loneliness involved not feeling they belonged in a social environment or community:

“No, I don’t feel part of this community, not at all…I feel lonely where I live, yes.”

[P51, female, White other, 46–55, urban]

“Sometimes the feelings of, well I do belong to that community but do I really? Am I a bona fide member of it? Feel like I am a hanger-on in the community.”

[P11, male, White British, 66–75, urban]

Some described a sense of thwarted belongingness, being ‘on the outside looking in’ or ‘missing someone or something, but not always knowing what.’

Not being able to find others who shared values and interests was part of the sense of not belonging:

“I feel like I struggle with my identity sometimes, like who am I and stuff like that. Yeah, I’m always like, trying to find like, my tribe, and like-minded people and stuff like that.”

[P29, female, Black, 26–35, urban]

For one participant this feeling of being ‘other’ and not belonging followed relocation from a large city to a more rural area with a different recreational culture:

“It’s not actually the act of being alone, it’s… not having [the right] people to do things with or to share things with.”

[P2, female, 36–45, White British, urban]

Social class played a role in some people’s lack of a sense of belonging:

“I’m from a working-class background, I’m very around middle-class professionals at work -who speak very differently to me. …I have to constantly think about how I’m coming across, how I’m speaking… makes me feel a bit disconnected…I rarely join them for lunch and things … I find it a bit like I’m an outsider.”

[P37, female, Black African, 26–35, urban]

Theme two: Connections between loneliness and mental health

Participants described close connections between mental health and feelings of loneliness. These included loneliness leading to deterioration in mental health, and feeling lonely because of impacts of mental health problems.

Theme three: Factors contributing to ongoing loneliness

As well as impacts of mental health problems, participants identified contributing factors to loneliness that included both historic root causes and current triggers, which were internal and external in nature.

Theme four: Ways of reducing loneliness

Two broad sub-themes emerged: external and internal approaches to reducing loneliness. These were categorised based on participants’ emphasis and experiences, but were often intertwined rather than being two clearly distinct categories. Strategies and ideas classed as “external ways to reduce loneliness” are those where participants emphasised how their experience depended on specific needs being met by people or environments external to themselves; those classed as “internal ways to reduce loneliness” highlight internal changes in thought patterns or other psychological shifts in participants’ ways of being and relating to the world around them.

Overall, different forms of social contact were seen as important for many participants in reducing loneliness, with quality of relationships and a sense of “belonging” being attributes of social contact that were particularly pertinent. Mental health challenges could act as a barrier to improving social contact and so addressing this was also seen as a key step for some participants. A booklet [31] (http://tiny.cc/Lonely) drafted by members of the research team summarises suggestions from study participants on coping with loneliness.

Main findings

Our findings conveyed how differently loneliness is experienced by different people, as expressed in terms of emotions, or even physical sensations. Loneliness appeared to comprise psychological elements, related to people’s thoughts and feelings about themselves in relation to other people and wider groups and communities, and social elements, related to the impacts of everyday interactions and contacts. Participants described a range of contributing factors to the origins of their loneliness, and the ways that their loneliness was maintained. They also perceived clear links between their loneliness and their mental health, and vice versa, and sometimes a feedback loop between the two.

Our participants’ accounts, as encapsulated in Theme 1, confirm that feelings of loneliness do not simply relate to dissatisfaction with the amount of time spent with others, but, perhaps more centrally, to not feeling connected to them in meaningful ways, and to not experiencing a sense of belonging. Loneliness has long been characterised as a physically and psychologically harmful manifestation of fundamental needs for social connection not being met [32]. Our findings regarding the importance of sense of belonging can be connected to investigations of loneliness from a social identity perspective, which supports that having multiple valued group memberships is associated with less loneliness and greater well-being [33]. The ways in which participants described both the nature of their loneliness and its origins were diverse, congruent with quantitative findings that suggest a complex causal web underlying loneliness [1]. Contributing factors identified included external factors such as losses and transitions or excessive time alone, but also aspects of personality such as introversion or lack of self-confidence, as well as the long-range impacts of trauma and adverse early experiences on the ability to form relationships (Theme 3).

Some of the themes illuminate aspects of loneliness relevant across populations, but a central aim of our work was to better understand loneliness among people living with mental health conditions. We found that loneliness and mental health appear intertwined in several ways (Themes 2). Participants described how feelings of not being connected to others could arise directly from a range of mental health conditions by pathways including feeling negative about self and others; withdrawing when depressed; and feeling unable to connect with others even when in company because of preoccupying social anxiety and hearing inner voices with negative content impeding trust and ability to socialise. This is in keeping with findings of greater loneliness associated with a range of mental health conditions [13, 34–36]. Some participants also described how their loneliness could lead directly to onset or exacerbation of the mental health conditions they were experiencing including depression, which is congruent with longitudinal studies establishing loneliness as an independent risk factor for depression [11, 12], and with findings of a bidirectional relationship between loneliness and depression [37].

Going beyond direct links from mental health symptoms to loneliness, the actions people took to cope with their mental health problems sometimes also placed them at greater risk of loneliness, such as when they withdrew from the stresses of social contacts and activities in order to recover and to ensure they had time and energy to cope with pressing practical needs. The circular relationship between loneliness and mental ill health identified in Theme 3 evoked the paradox identified by Achterbergh et al. [21] in their meta-synthesis of qualitative studies on loneliness and depression among young people: social withdrawal to cope with mental ill health can result in loneliness that further exacerbates mental health problems.

Significant contributors to loneliness among people with mental health problems in our sample were stigma and self-stigma associated with mental health problems, especially as impediments to having a sense of belonging. This is in keeping with previous findings of an association between self-stigma and social withdrawal among people with severe mental health problems [38], and of high levels of persisting mental health stigma despite a longstanding UK public anti-stigma campaign, focused primarily on common mental health problems [39].

Experiences of stigma and social exclusion related to mental health intersected for many participants with social exclusion associated with being part of other disadvantaged or marginalised groups at increased risk of mental health problems, including racial and sexual minorities and people with disabilities, and with impacts of social deprivation. As Lever-Taylor et al. [40] argue in their qualitative study of perinatal women, an intersectional focus is helpful in understanding social drivers of loneliness. Participants’ accounts of the many impediments to a sense of belonging, including stigma, reinforce a need to take a societal and community as well as individual approach to understanding loneliness. Thus, loneliness appears to result not only from individuals’ inability to connect, but also from failures of communities to welcome and integrate people living with mental health problems, and from practical barriers to connecting with others that result from poverty [41].

Many of the themes and sub-themes so far discussed cohere with previous literature, for example illuminating potential mechanisms underpinning epidemiological findings. To our knowledge, our exploration of the strategies and types of help people employ to combat loneliness is novel (Theme 4). We found that many participants were aware of their loneliness and its impacts, and of a variety of strategies that could help: we reflect further on implications for interventions below.

Strengths and limitations

Our study represents a substantial contribution to the limited qualitative literature on loneliness among people with mental health problems. We recruited a diverse sample, encompassing a range of backgrounds, types of mental health problem, service use histories and locations. Although using a digital platform for most interviews will have excluded a substantial section of the population using mental health services, we did conduct some interviews by telephone and face-to-face to accommodate the digitally excluded. Lived experience was embedded in the study at each stage, with people with relevant personal experiences designing interview guides, conducting interviews, analysing data and writing up findings.

Limitations include that we conducted a broad-brush analysis of a large sample of interviews, and searched for commonalities across a group that was very diverse in characteristics and experiences rather than focusing in more depth on more defined groups. Some people discussed links between specific mental health conditions and onset or exacerbation of loneliness, but our sample was very varied and diagnosis was based on self-report, so we cannot discuss links between particular symptoms and conditions and loneliness in depth; a potential direction for future work. We did not measure severity of mental health difficulties, and note that while a majority had used specialist mental health services (inpatient or community teams), thirteen participants reported not having used services for mental health.

The majority of interviews were conducted during the COVID-19 pandemic, and we have reported baseline [23] and follow-up [26] findings elsewhere, but the pandemic context may have influenced accounts of loneliness, even though participants were encouraged to talk about experiences of loneliness in general rather than specifically in a pandemic context, and appeared for the most part to be doing so.

Research and clinical implications

The rich and complex accounts given by participants of the nature of their experiences of loneliness, and the great variety of pathways into and out of it, indicate considerable further scope for research to understand associations between loneliness and mental health. In qualitative research, many of our themes warrant more in-depth exploration, including focusing on particular groups at high risk of loneliness, or who are currently lonely. In quantitative research, it would be valuable to investigate further the longitudinal relationships between loneliness and mental health problems, and the extent to which contributory factors identified in our study are also reflected in epidemiological analyses. Much research on loneliness has employed single-item or brief measures not tailored to people with mental health conditions: we reflect that such measures are unlikely to capture the diversity of experiences of loneliness and links to mental health.

As yet there are very few clearly evidence-based strategies for helping with loneliness among people living with mental health problems [16]. Our findings show people with mental health problems recognising and taking steps to address their loneliness. This supports the greater deployment of co-production approaches, incorporating the forms of help that people with relevant personal experience see as potentially effective. With few exceptions [42], interventions tested thus far have tended not to be co-produced. The diversity of pathways into and out of loneliness that people described, and of suggested strategies, indicate that a variety of approaches are potentially helpful depending on the nature and context of loneliness, delivered singly or in combination. These would include self-help and psychoeducation about day-to-day strategies that people have found helpful, psychological interventions focused on thoughts and feelings about others, and social approaches to help people develop meaningful connections and a sense of belonging [15, 43].

The range of mental health-related factors triggering and perpetuating loneliness suggest benefits to developing or adapting specific strategies for this population rather than deploying strategies developed for the general population. Approaches to loneliness in a mental health context based on peer support seem rarely to have been reported, but have clear potential benefits such as in overcoming obstacles due to stigma and self-stigma, fostering a sense of belonging and supporting people with self-help strategies to reduce loneliness. Finally, our perspective in this study was on improving the support offered in mental healthcare settings, but our findings illustrated the community-level, intersectional and socio-economic drivers of loneliness in multiple ways. We suggest that addressing such drivers will have a central influence on whether levels of loneliness can be reduced among people living with mental health problems.

Lived experience commentary written by Beverley Chipp

This paper must be understood in the context of the COVID-19 lockdown occurring when we had only conducted a few interviews. All participants had already felt lonely as this was our inclusion criteria, but limitations upon movement and social contact imposed by government, and, significantly, the cessation of many support services will have influenced subsequent interviewees’ responses. For some these changes made things worse and for others life became easier. The shifts triggered significant reflections on life and participants’ relationships, and perhaps provided richer data than would have been collected in normal times.

Access to mental health support has been increasingly difficult over the last decade but the suspension of face to face services, or for some, services altogether, exacerbated loneliness leaving many feeling brushed aside.

We learned that loneliness and social isolation were distinct entities and experienced in multiple and diverse ways. The causes and perpetuating factors were more complex than we had imagined, even though we identified with some of the narratives.

Solutions need to take a psychologically informed, holistic approach rooted in community and co-production. Without addressing stigma and other barriers within society–prejudice and discrimination, hostility against those that don’t ‘fit’–people’s sense of belonging will be hindered, and loneliness increased. Where loneliness has childhood origins it may be more difficult to unpick. Simply engineering people to be with other people is too simple a solution for most individuals also living with mental health conditions.

Commonality between lived experience researchers and participants fostered greater trust in interviews and thus sharing of deeper insights. We felt that research often stays within a ‘research bubble’, not reaching the people it directly affects. Recognising the wealth garnered from asking participants ‘what helped?’ and their willingness to offer suggestions, we co-produced a self-help booklet (http://tiny.cc/Lonely) which acknowledges their resourcefulness.