Understanding self-harm in older adults: A qualitative study.

BACKGROUND: Self-harm is the leading risk factor for suicide, with elevated rates reported amongst older populations. This study explores how older adults experience self-harm, identifying factors leading to self-harm.
METHODS: Semi-structured interviews with older adults (≥ 60 years) engaging in self-harm and support workers from third sector services in England. Older adults were invited to participate in a follow-up interview. Interviews were recorded, transcribed verbatim and data analysed thematically. Ethical approval obtained from Keele University's Ethics Review Panel. A Patient Involvement group contributed to study design, data analysis and interpretation. OUTCOMES: Between September 2017 to September 2018, 24 interviews were conducted involving 16 participants: nine older adults and seven support workers. Eight older adults consented to follow-up interviews. All older adults reported diagnoses of mental illness in addition to physical illness. Participants identified diverse stressors accumulating over the life-course leaving older adults particularly vulnerable to self-harm. Such stressors included adverse events, loss, interpersonal and health problems. A sense of shame and stigma amongst older people using self-harm to manage distress was also reported.
INTERPRETATION: Self-harm is often concealed due to stigma and shame, being further accentuated amongst older adults, which may result in low levels of medical help-seeking behaviour for self-harm. Self-harm occurred along a spectrum of no-suicidal intent to high-levels of intent, suggesting self-harm holds different functions to older adults. Clinicians should be aware of the existence of self-harm in this age-group, and the heightened risk amongst those with comorbidities so adequate assessment, support and/or referral is provided.

Research in context

Evidence before this study

UK's National Institute for Health and Care Excellence (NICE) Guidelines (CG133) defines self-harm as “any act of self-poisoning or self-injury carried out by a person, irrespective of motivation”.

Before conducting this study, we undertook a systematic literature review of characteristics (including clinical, risk factors and lived experiences) of self-harm in older adults. 40 articles (n = 62,755 older adults) were included. Self-poisoning was the most commonly reported method (86%), and physical illness was high (40%) amongst older adults. Risk repetition (17%) was high, with increased risk amongst older adults with self-harm history, previous and current psychiatric treatment. Only three studies used qualitative research methods to explore motivations for self-harm in older adults. Loss of control, increased loneliness and perceived burdensome ageing were reported self-harm motivations. However, these experiences were limited to the context of self-harm with exclusive suicidal intent. Given the dynamic role self-harm motivations have amongst people who self-harm, it is important to explore the different experiences and motivations for self-harm (suicidal or non-suicidal) as this would aid in clarifying the heterogeneous terminology used to refer to self-harm and further understand experiences of self-harm in later life.

Added value of this study

To our knowledge, this is the first study using qualitative methods to explore the experiences and reasons of self-harm in older adults, regardless of suicidal intent, as well as perspectives of third sector support workers. Findings from this study show that self-harm is concealed in older adults given a sense of shame and stigma experienced within this population, which may lead older adults not reporting their self-harm or seeking support. This study contributes to existing arguments from other qualitative studies which have already begun the job of unsettling the relationships that are claimed between self-harm and future self-harm and suicide. This study is the first conducted with an older adult population that confirms that the relationship between self-harm and suicide is more complex than previously conceptualised, given that some older adults report engaging in self-harm behaviour to avoid suicide. This is an important finding for clinicians to consider when assessing older adults who self-harm.

Implications of all available evidence

Current national and international clinical guidance highlight the importance of identifying and managing older adults with self-harm behaviour, due to the increased risk of repetition and suicide. Our findings highlight self-harm in older adults is often hidden, which may result in low levels of medical help-seeking behaviour. Self-harm holds no singular function for older adults, however clinicians should be aware of the existence of this behaviour in later life, and the heightened risk amongst those with mental and/or physical health conditions. Furthermore, it is important for clinicians supporting older adults who self-harm, to engage with older adults in order to understand their motivations to self-harm to provide appropriate support and/or referral.

Alt-text: Unlabelled Box

Introduction

Self-harm, as defined by NICE guidelines (CG133), is “any act of self-poisoning or self-injury carried out by a person, irrespective of motivation” [1], constitutes urgent public health challenges worldwide. Self-harm is the major risk factor for suicide, which has highest rates amongst older adults [2]. Nonetheless, self-harm is also presented in the absence of suicidal intent (i.e. non-suicidal self-injury). Amongst older adults, self-harm is more common in women, people with physical and/or mental health conditions and those with previous self-harm history [3], [4]. When compared to younger populations, self-harm rates are less prevalent in older adults. However, self-harm rates might be under-estimated because of shame and perceived or felt stigma [5], [6], lack of disclosure and difficulty working with older adult populations [3], [7]. Furthermore, amongst older adults, evidence suggests there to be increased lethality associated with self-harm [3]. This increased lethality often results in higher resource and treatment costs due to medical complications caused by frailer health [8].

Current self-harm and suicide guidance suggest multi-factorial approaches to prevention and management [9], [10]. These approaches include engagement from healthcare systems, but also education and social settings, and overall community given the different influencing factors to self-harm and suicide. Alongside contact from healthcare services, increasingly third sector organisations, also known as voluntary or community organisations, are supporting people with mental health problems [11], including self-harm [12], filling in the gap of limited service provision from statutory services. Amongst the older adult population, the third sector has a crucial role in care-coordination and delivery of social and healthcare services [13]. Several studies have reported repeated contact of older populations and people who self-harm with the third sector [13], [14].

A robust evidence base is key to prevention and management. Whilst current evidence suggests a strong link between self-harm and increased suicidal intent amongst older adults, this evidence is based predominantly on quantitative research designs which only offer a partial understanding of this complex phenomenon [2], [3]. Moreover, the majority of research has been conducted in clinical settings (predominantly hospital based) limiting their applicability to community samples [3], [4]. Understanding lived experiences of older people who self-harm is crucial to developing effective interventions. The few qualitative studies that have been conducted with older people are limited to a self-harm definition of exclusive suicidal intent (e.g. suicidal attempts), and the relationship with suicidal intent and self-harm in older adults remains unclear [3]. Further exploration and understanding of the spectrum of suicidal behaviour and self-harm in older adults is needed. The aim of this study was to (1) explore the diverse narratives of self-harm experiences in later life from the perspectives of older adults with self-harm behaviour and third sector workers; and (2) identify how older adults and third sector workers give meaning to the factors leading older people to self-harm.

Methods

Study design

A qualitative methods approach was adopted using semi-structured, in-depth interviews to allow participants to extensively narrate their accounts, while enabling the interviewer to guide the interview with predetermined topics [15], [16], [17]. Third sector workers were interviewed once, whilst older adults were invited to consent to one follow-up interview. This two-stage approach with older adults was chosen because of the sensitive nature of the topic, give time to establish rapport between interviewer and participant and provide the opportunity to develop a deeper understanding of individuals' experiences [18]. Follow-up interviews were spaced approximately one-month apart to allow for identification of emerging themes. The participant was then invited to reflect on and expand/clarify as they wished in the subsequent interview. In providing time for reflection on their first interview, the intention was also to support participants in having control over the focus of the second interview and the data they contributed to the study.

Patient and Public Involvement and Engagement (PPIE)

Patient involvement is increasingly used in health research for its potential to enhance quality and appropriateness of research, and increase the acceptability and implementation of interventions [19]. In light of this, a PPIE group was established with the aim of contributing to the development and refinement of recruitment strategies, analysis of transcripts, identification of themes and dissemination- see Troya et al. [20]

Participant recruitment

Two participant groups were involved: (1) older adults (aged ≥ 60 years) with current and/or recent (within being the age of 60) self-harm history; (2) third sector support workers with previous experience supporting older adults who self-harm. Table 1 expands further on older participants' current or previous engagement with self-harm behaviour. Participants from community settings were recruited through the different avenues summarised in Fig. 1. Purposive sampling was applied. Participants were excluded from taking part in the study if they lacked capacity to consent or could not understand or speak in English. For further information regarding steps followed for participant recruitment, see Troya et al. [21]

Table 1

Characteristics of older adults (1.1) and support workers (1.2).

1.1 Characteristics of older adults
	Gender	Age	Health conditionsa	Psychosocial context	Self-harm history	Support
F1	Female	62	• Personality disorder • Fibromyalgia, diabetes, heart disease, scoliosis, arthritis	• Early start of mental health difficulties with child sexual abuse experience • Living alone, experiencing loneliness • Family history of mental illness • Experienced loss of children in adulthood • Limited mobility due to health	Start: early teens.On and off engagement of self-harm	• Long history of overdose hospital admissions • Limited family support • No longer attending group for self-harm • Medication seen by General Practitioner (GP) • Infrequent contact with Community Psychiatric Nurse (CPN)
F2	Female	72	• Depression and alcohol misuse • Irritable bowel syndrome, arthritis, pancreatitis	• Childhood sexual abuse and overall stressful upbringing • Interpersonal difficulties with family & family history of alcohol misuse • Early retirement due to mental health conditions	Start: early childhoodNo longer engages in self-harm behaviour	• Regularly attends sector group for self-harm • Previous contact with secondary care • Medication overseen by GP
F3	Female	60	• Personality disorder • High blood pressure, hypothyroidism	• Multiple childhood stressors including sexual abuse, bullying, neglect, encounters with judiciary system • Family history of mental illness • Partner bereavement leading to experience of loneliness • Full time factory worker	Start: early teensOn and off engagement of self-harm	• Regularly attends sector group for self-harm • Medication overseen GP • Support received by CPN • Long history of contact with primary and secondary care services
M1	Male	67	• Depression • High blood pressure, heart disease	• Early childhood experience of shame caused by secrecy of adoption • Experienced multiple losses of family members and friends • Health conditions disrupted life since early 40s leading to job loss • Previous experience in research as a participant	Start: 40sNo longer engages in self-harm	• Multiple experiences with counsellors and contact with primary and secondary care • Regularly attends sector group for self-harm • Medication overseen by GP
F4	Female	65	• Personality disorder, eating disorder & post-traumatic stress disorder • Osteoporosis, high blood pressure	• Early childhood experiences of sexual abuse, violence and neglect • Family history of mental illness • Previous experience in research as a participant	Start: 40sOn and off engagement of self-harm	• Recently joined third sector self-harm group • Support from primary & secondary care services • Receives family support
F5	Female	62	• Depression • Fibromyalgia, sciatica, ankylosing spondylosis, arthritis	• Early childhood loss with death of mother, adopted by grandparents • Ongoing conflict with husband & interpersonal problems with neighbours • Multiple bereavements • Health conditions affecting mobility and everyday life	Start: 60Currently still engaging in self-harm behaviour	• Recently joined third sector self-harm group • Previously received support from psychiatrist • Medication for depression overseen by GP • Receives family support
M2	Male	61	• Pica, anxiety & post-traumatic stress disorder • Dystonia, heart failure, diabetes, liver disease, ulcerative colitis	• Limited mobility due to health conditions, affecting everyday life activities • Several negative experiences with health care facilities • Loss of family members • First time talking about self-harm	Start: 40sCurrently still engaging in self-harm behaviour	• Previously support by counsellors • Currently sees psychiatrist • Attends third sector group for people with dystonia • No support for self-harm
F6	Female	62	• Depression, eating disorder • Arthritis, walking disability	• Ongoing experience of violence and abuse • Escaped violent and life-threatening marriage, forcing her to re-locate cities • Has lived with self-harm for over 50 years • Identified as ethnic minority, highlighting difficulty to access support	Start: Early childhoodOn and off engagement of self-harm	• Received limited support for self-harm • Attends third sector self-harm group • Soon to start seeing a private counsellor
M3	Male	60	• Eating disorder, obsessive compulsive disorder (OCD), Personality Disorder • Anaemia	• Divorce of parents experienced in childhood • Self-identified as homosexual • Recently moved to England from the United States • Highly educated and with previous experience in research • High levels of insight and self-awareness	Start: Early childhoodCurrently still engaging in self-harm behaviour	• Talking therapy received for over a decade but stopped once moving to England • Attends service user group and is on the waiting list for a self-harm hospital programme • No family support

As reported by participants.

As identified by participants.

Fig. 1

Flowchart of recruitment methods (from Troya et al., [21]).

Previous self-harm history

Started the only self-harm group in [city in North West England] since she saw the lack of support for people who self-harm

Has only recently started the group and supported people who self-harm

Looking for funding in order to make group a third sector organisation for people who self-harm

No mention of self-harm or mental health history

Has worked with vulnerable populations such as supporting abused women in the past

Majority of experience supporting people who self-harm through observation of previous groups' support worker

Previous self-harm history

Working as a support worker for many years but only recently with people who self-harm

Looking to receive further training for supporting people who self-harm

Traumatic experience in teens (rape) which led to self-harm

Received support from family and local third sector group for self-harm in the last decade

Took over support worker/facilitator role in the group 11 years ago, while still in recovery for self-harm

Has completely stopped self-harming for 8 years

Multiple experiences with research projects as well as being a lay board member for various local suicide prevention boards

No mention of self-harm or mental health history

Support offered to older adults focused on social services benefits and overall wellbeing

Has not received training on how to manage and deal with people who self-harm

Has trouble relating and understanding self-harm in members.

Previous counselling/psychology background

Previous experience in research with older adult's population and mental health

Previous history of self-harm and suicidal attempts from early adolescence due to childhood sexual abuse

Only received support for his mental health when diagnosed with cancer in his adulthood years

Started facilitating group after volunteering at service user led group given the need of setting up a self-harm group

Participated in other research projects and actively involved in public speaking and mental health and self-harm awareness

Ethics

Ethical approval was sought and obtained from the Ethics Review Panel at Keele University (Ref: ERP1333). After expressing interest in the study and confirming eligibility criteria, participants were approached and supplied information sheets explaining details of the research. All participants had the opportunity to ask questions prior to signing the consent form. Informed consent was obtained at the start of each interview, and checked again at the end. Consent was recorded verbally in the case of telephone interviews. In the case the participant agreed to take part in a second follow-up interview, informed consent was again separately obtained at the time of that interview.

Data collection

Participants were recruited from September 2017 to September 2018. Once eligibility criteria were confirmed and informed consent obtained, interviews with participants were arranged at a time and venue preferred by each participant, and telephone interviews were offered in addition to face-to-face. In addition to the topic guide, which included areas such as self-harm history, motivations and access to care, socio-demographic data were also collected.

Interviews were conducted by the first author (IT) at participants' preferred location (Keele University, participant's home, or third sector venue), or at University facilities when conducting telephone interviews. Interviews were digitally recorded and transcribed verbatim by the first author. A reflexive diary was kept in order for IT to annotate thoughts and reflections before and after each interview.

Analysis

Data were analysed using thematic analysis and constant comparison methods [22], [23], following an iterative process, where each interview was coded before new data generation started. This process allowed for initial codes to be further explored and analysed whilst new interviews were being conducted and analysed. When data collection and analysis indicated no new themes emerging, thematic saturation was considered reached and recruitment ceased [24]. Thematic saturation is defined as the criterion used in qualitative research for deciding when to stop data collection once no new themes are emerging [24]. Thematic saturation was reached in both participant groups.

IT led on the coding and followed the steps described in Fig. 2 for data analysis. As further described in Fig. 2, after coding, creating and revising the initial themes, themes were compared and contrasted across both datasets (of older adults and third sector workers) to integrate the overall findings. All transcripts were coded by at least two of the authors and discussed amongst all authors.

Fig. 2

Summary of Data analysis process (from Troya et al., [21]).

Unlike other methods of qualitative data analysis, thematic analysis has no specific theoretical or epistemological approach, making it flexible to the researchers' stance and study questions to be addressed [15]. Analysis, alongside identification of key themes, were influenced by the research team's clinical and academic backgrounds, which consisted on a mix of medical sociologists, critical gerontologists, academic GPs, clinical psychologists. Also influencing data analysis and interpretation was the input received from the PPIE group. The involvement of diverse researchers' backgrounds and PPIE input contributed to the trustworthiness of the data [25], [26].

Role of the funding source

The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the manuscript. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Results

Twenty-four semi-structured interviews were conducted involving sixteen participants (nine older adults and seven support workers). Follow-up interviews conducted with eight older adults. Interviews lasted from 28 to 129 min. Sociodemographic and other personal characteristics of participants are summarised in Table 1, with all older adults reporting both physical and mental illness, and over half (n = 4) of the support workers reporting previous self-harm history. Overall, findings followed a life-course perspective where older adults' lives, and in particular their mental health, were composed and impacted by the accumulated lived experiences and social structures formed from early childhood to later life [27].

Stressors contributing to self-harm

Several stressors were identified as influencing older adults' self-harm throughout the life-course, including health problems, adverse childhood events, interpersonal problems, loss and loneliness (see Table 2). The accumulated impact of such stressors over the life-course was reported and interpreted by participants as leaving older adults vulnerable and engaging in self-harm in order to deal with distress.

Table 2

Stressors for self-harm in older adults.

Stressors	Data from older adults	Data from support workers
Health problems	“Me mental health issues I've had since I was 15. I took my first overdose when I was 14 and it's gone on from then […]Since 1985 I've had physical health problems that have just escalated until now that I'm practically housebound without an electric wheelchair.”-F1, older adult	“Yeah cause mental health problems can be a big one. You know you very rarely get someone who self-harms that doesn't have a mental health issue on the side. And sometimes it's getting that person to realise there is something wrong with their mental health.”-M5, support worker
Adverse childhood events	“So it wasn't just problems with my dad, it was problems with my grandad, some sexual abuse. And uh with my dad it was a lot of physical and emotional abuse. I don't know, there might be other things that went on as well but I don't know, I find those hard to look at.”-F4, older adult	“There's a majority of people who have suffered quite significant trauma who attend [organisation name #1]. And there's a high number that have experienced uh childhood sexual abuse.”-F8, support worker
Interpersonal problems	“I was fleeing violence from a very violent husband and the police sent me to the hostel (.) He attempted to try and kill me a couple of times so it was for my own safety really. During this time I just self-harmed in any possible way I could because that's what I knew how to do best.”-F6, older adult	“For instance that lady taken an overdose uh she disclosed to me that it was after she had been in a marriage of years of abuse and maybe domestic violence, emotional violence and her husband was a drinker as well. So uh that's why you know she was uh she was depressed, anxious and very emotional over the phone.”-M4, support worker
Loss, bereavement and loneliness	“The only the only time I've taken an overdose of morphine and that's that was when me dad died in the July.” –M1, older adult	“You know I was obviously sorta probing quite a lot on how it was affecting him and when this particular gentleman he uhm (.) He had come home from work to find his mother dead on the floor on the home that we were in. So it was obviously like a grief reaction. He wasn't (.) Uhm he told me during the interview he couldn't even attend his mother's funeral, he couldn't leave the property, so it seemed pretty instant. (…) So yeah the loss of his mom yeah yeah so that's how it come about.”-F11, support worker

Self-harm motivations

In addition to being used to deal with distress, participants identified various motivations for self-harm in this population, including: a cry for help, a coping mechanism and to regain control.

Cry for help

Participants reported self-harm as used instrumentally by older people, either consciously or unconsciously, to seek help and support, with some using self-harm to avoid suicide:

“I don't think it's suicidal, it's a cry for help. I've never really uh well I only went out to try to kill myself twice but otherwise my self-harm hasn't been associated with suicide. Like a soothing blanket, even when you're cutting, it's soothing, it gives you that comfort that you're not getting from other people.” –F6, older adult.

Older adults identified their self-harm as a ‘cry for help’, reflecting the self-harm episode as one that goes beyond bodily expressions of distress, seeking help for their distress through self-harm. Furthermore, the relationship with suicide was explored amongst participants, with older adults experiencing self-harm as a coping mechanism rather than a suicidal expression. In contrast, despite many having experienced self-harm personally, support workers used the more pejorative term of ‘attention seeking’.

“You got people that self-harm to get attention. That sounds awful but I've seen some of them.” –M5, support worker.

The different views regarding self-harm as a ‘cry for help’ and ‘attention seeking’ have the potential of holding negative implications in clinical and caring settings such as those provided by support workers. In response to this, conceptualising self-harm as ‘attention seeking’ left some older adults feeling their desperation was not understood.

“I don't know what people think we're doing. Ok we might be attention seeking, I don't know, maybe as well there's something else. But if that's what people are doing and there's nothing for attention, no support, they're still gonna hurt themselves more and end up in hospital. And it's only gonna cost the NHS more in the end. So why can't they just listen to them at the time and say what is it?”-F4, older adult.

Coping mechanism

Self-harm was identified as a coping strategy to deal with stressors accumulated throughout the life-course. These stressors often left older adults in a vulnerable position, feeling the need to self-harm to manage distress as no alternative coping mechanism was found.

“The reason you do it, when you self-harm when you cut yourself, it's because the pain you've got and it brings it out of you instead of holding it in. I know you're hurting yourself but it's letting the stress out in one way.”-M1, older adult.

Older adults gave meaning to their self-harm as a coping mechanism to deal with adverse events and pain as reflected in the previous account. Whilst for some the use of self-harm was a recently acquired coping mechanism, older adults experienced earlier traumatic events as leading to early manifestations of self-harm.

“That's why I started self-harming. The abuse and everything else. It was too much to handle.”-F1, older adult.

Older adults experienced such adverse childhood events as influencing their self-harm in later life. But not all older adults started self-harming from a young age. Others used self-harm to deal with more recent events which caused them distress and found comfort using this behaviour.

Furthermore, engaging in self-harm, both through overdosing and self-injury as reported by participants, allowed older adults to escape emotions which seemed unbearable.

“So all them little things collected up and I just wanted die. I just thought there is no getting away from anything, so I just wanted to die. It was a way of getting out of it.”-F5, older adult.

‘Numbing pain’ was commonly reported amongst older adults. Older adults experienced the action of self-harm as allowing them to momentaneously numb the psychological pain felt by the different accumulated events experienced throughout the life-course. Once again, this numbing of pain was described as being reached through diverse engagements of self-harm, including both overdosing and self-injurious behaviour (e.g. cutting).

“I started self-harming because of the pain, not having anyone to talk to about it, I didn't know what was going on, this turmoil.”-F6, older adult.

Regain control versus gratification

Some participants reported feeling a loss of control over their lives, and feeling powerless. Support workers with previous self-harm experience were also able to offer insight into the understanding of why older people self-harm, based on professional and personal experience with self-harm.

“Behind self-harm there's pretty much one reason…and that's all about getting control or regaining some sorta level of control, even if it is short-lived. We feel that the situation or life in general is out of our control. And by making a decision to hurt ourselves we're just gonna put ourselves in the driving seat for a bit.” –F8, support worker.

In a sub-set of participants, self-harm not only resulted in regaining control, but was also a sense of gratification:

“It's also kind of like an achievement (.) Successfully cutting or creating a scar or incision, for myself it's powerful. Yes, you feel the pain but as kind of cliché as it is, it's kind of like opening up a balloon. There's an element of accomplishment, success, gratification and reward.”-M3, older adult.

Other participants describing feeling gratification from their self-harm, stated self-harm produced a mixture of pleasure and pain.

“It hurt but it was satisfying at the same time (..) I also think it gives me nano seconds of pleasure when I do it […] it's sorta like an exquisite pain.”-M2, older adult.

Whether producing a sense of comfort because of regaining control or leading to feelings of gratification, self-harm allowed older adults to obtain these positive outcomes, even if they were momentary. In addition to the variation in motivation, it is important to note how self-harm could be used by older adults to avoid suicidal behaviour and even deterred them from more severe suicide attempts. This varied throughout the life-course.

“She literally put herself in A&E every Friday night so they would keep her in during the weekend so that she couldn't kill herself. She never took enough that would kill her, but took enough that they'd keep her in. She worked out that was her way of surviving, when everyone else was thinking she was trying to kill herself, she was just like no I'm not, I just wanted someone to help me.” –F10, support worker.

“When I was a kid, I think it's pretty safe to say it wasn't about killing myself. As an adult when I was cutting my wrists […] I thought cutting myself like I did, then you would die. Then I did the whole pill thing which then obviously didn't work. And then I went back to cutting on the wrist but not for suicidal but to kind of like feel the pain and see the pain.”- M3, older adult.

Secrecy, shame and stigma

“It's the stigma around it. When they're in their 60s and 70s there might be a bit of shame in telling people what they're doing. And for the grandma or grandad to turn around and see scars in your arms you know it's stigmatising […] They're the role models aren't they? Who you look up to. They're the strong ones.”-M5, support worker.

Secrecy, shame and stigma related to self-harm emerged from participants' accounts, and described as more accentuated in later life:

“The older somebody gets, the harder it's to talk about it. They think I shouldn't be doing this silly behaviour, everyone will think I'm just silly […] some think I'm too old to start this, that's like teenage behaviour.” –F7, support worker.

Participants reported feeling shame as a result of their behaviour, which resulted in felt stigma. For some, higher levels of shame and stigma were experienced because of social and family expectations associated with this age group, resulting in secrecy.

“I felt embarrassed because of my age. […]I'm thinking its girls that only should, 16, 17 year olds and they're self-harming and here is me, I should know better.” –M1, older adult.

However, societal changes regarding attitudes towards self-harm and mental health problems were also reported. These societal changes were experienced as aiding older adults who had been living with self-harm for several decades and could reflect on the attitude changes:

“I think it's been a long journey, but I think I'm in a fortunate position now and people are much more open to mental health problems and self-harm, uhm, I think the attitudes have changed anyway, it's more common now. I think I think I've seen changes in people attitudes in the last 20 years.” –F3, older adult.

Furthermore, although there were only two participants from minority groups (Lesbian, Gay, Bisexual, Transsexual-LGBT groups and ethnic minorities), participants' reports seem to suggest stigma was further accentuated amongst these groups.

“Black communities don't talk about these things. This whole conversation wouldn't be happening in a Black or Asian family. It's the decency, you've gotta keep decency. So it's hard for them to get out because the family and community don't want them out.”-F6, older adult.

“Within the gay community, an open cut you know because of HIV it's negative. You know [laughs] gay people are afraid of blood because of it. So yeah so that brought it to another level.” –M3, older adult.

Discussion

This study is the first using qualitative methods to offer an in-depth exploration of the reasons and experiences of self-harm in older adults, regardless of suicidal intent. A conceptual model (Fig. 3) was developed, with a PPIE group, to illustrate the process of self-harm as experienced in later life. Older adults experienced different stressors throughout the life-course: adverse childhood events, loss, health and interpersonal problems. Self-harm was often expressed as a ‘cry for help’. Participants gave meaning to older adults' self-harm as acting like a coping mechanism in order to: 1) numb pain; 2) escape reality; 3) avoid suicide; 4) express intense emotions; 5) deal with traumatic events. The consequences of self-harm as a coping mechanism often led older adults feeling a sense of regaining control, and in a sub-set of participants gratification.

Fig. 3

Motivations for self-harm in older adults within a suicidal spectrum.

Previous research exploring motivations for self-harm in older populations explored different population groups by adhering to a broad definition of self-harm (e.g. inclusion of indirect self-harm) [28]. Our study offers a different perspective by adhering to NICE guidelines (CG133) definition of self-harm which includes direct self-harm only, excluding indirect self-harm (e.g. refusal to eat), given the distinctive nature of the two. However, differences between indirect and direct self-harm may result in different levels of experienced shame and stigma, therefore future research is needed to explore such differences.

Findings from this study add to the previously identified motivations and functions of self-harm in older adults, in addition to expanding the range of self-harm behaviour in older adults (non-suicidal self-harm and attempted suicide) [3]. As other authors have previously conceptualised [29], [30], findings support the theory of self-harm and suicidal behaviour being more than unidimensional constructs that can be understood through singular behaviour processes. In this study, self-harm was presented within a spectrum of suicidal behaviour, with some episodes being internalised as a suicidal attempt, and others as having no suicidal intent (see Fig. 3). Older adults' experiences with self-harm varied within the spectrum, with certain episodes holding no suicidal intent and others being self-identified as attempted suicides; these experiences changed across the years as self-harm held different functions to older adults. This is an important finding given the lack of clarity within the literature in understanding self-harm functions in later life.

Through the use of qualitative methods, this study provides contributions and lived-experience insights to research and clinical practice regarding the commonly assumed understanding of self-harm, particularly the relationship with suicide. Previous research has identified self-harm history as a risk factor for self-harm and subsequent suicide [3], [4]. Furthermore, previous research suggests that the relationship between self-harm and suicide in older adults is unambiguous [3]. However, findings from this study suggest that the relationship between self-harm, self-harm repetition and possible suicide is more complex given that in some older adults, engaging in self-harm behaviour allowed suicide avoidance. This study contributes to existing arguments from other qualitative studies conducted with younger populations [31], [32], [33], [34], [35] which have already begun the job of unsettling the relationships that are claimed between self-harm and suicide. This study is the first conducted with an older adult population that confirms that the relationship between self-harm and suicide is more complex than previously conceptualised. Furthermore, similarities between this research and previous research conducted with younger populations exploring self-harm motivations suggests some aspects of self-harm may be resilient across the life-course. Clinicians may wish to take into consideration these findings when assessing older adults who self-harm in order to look for safe-management of self-harm.

Within medical sociology, stigma is a well-known concept used to describe an attribute or behaviour (e.g. self-harm) which can be socially discrediting, by being socially undesirable [6], [31], [36]. Within the self-harm and suicide literature, stigma is also a well-known concept and previously reported, attributed to the high levels of shame reported by those who self-harm [6], [31], [36]. Older adults' description of self-harm being concealed, and secretive behaviour is congruent with the literature of stigma and self-harm. Furthermore, traditionally, self-harm is conceptualised as being a younger person's concern as participants highlight. Self-harm in later life may accentuate levels of shame and felt stigma (defined as the feelings of shame arising from the fear of being discriminated by others [37]) even further because of social expectations around older people [37]. Feelings of shame and stigma may result in older people being even less willing than younger populations to disclose their behaviour, which in turn may be leading to the under-estimation of self-harm in older adults-the documented ‘iceberg model of self-harm’. [7] Felt stigma and shame can lead to decreased help-seeking as has been seen in other populations experiencing mental illness [31], and even more accentuated in minority groups such as ethnic minorities and LGBT groups [38], [39]. Findings from this study confirm the accentuated experiences of shame and stigma amongst older ethnic minorities and LGBT groups, however they were based on reports from two participants, therefore further research with older minority groups is needed to confirm such findings.

There are three main limitations to this study. The first is regarding the fairly homogenous sample. Diverse recruitment strategies, informed by PPIE discussions, were taken in place in order to identify a heterogeneous sample and thus included recruiting participants from community settings. However, the majority of participants (n = 12) came from one recruitment avenue: self-harm support groups. We believe this was due to older adults attending such support groups felt more comfortable disclosing their self-harm to others, as opposed to other potential participants from the alternative recruitment avenues which may have not disclosed their self-harm before (as was the case with older adult M2, only participant recruited from newspaper ads). In addition to the diverse recruitment strategies, two ethics amendments were sought and obtained (June and November 2017) in order to expand the recruitment avenues given paucity in recruitment. The first amendment consisted on expanding the study location to England rather than being limited to the West Midlands, while the second amendment consisted on adding advertisement through local media such as newspaper ads. These amendments did indeed result in further participant engagement in the study. However, despite efforts to include a diverse sample, older adult participants were confined to a younger age group (mean age: 63.4; SD = 3.72) which may limit the transferability of findings to other later life cohorts. Furthermore, participants were predominantly White British, meaning these results may not be applicable to ethnic minorities. Research conducted with older ethnic minorities may have resulted in further themes in analysis. Although this study received first-hand experiences of self-harm as experienced in older adults, these results are limited to older adults and third sector workers perspectives. Interviewing other participant groups such as healthcare practitioners or carers would add other important perspectives to self-harm in older adults. Lastly, the relatively small sample size of each of the participant groups may be seen as a limitation. However, it is worth noting that amongst the included participants, different accounts were provided, and data collection stopped once saturation of data was reached in both participant groups, with no new themes emerging.

Despite these limitations, participants reported different experiences with self-harm in later life, through different onset of self-harm, as well as diverse clinical diagnoses and support. Furthermore, our study also explored self-harm in older people from the perspective of third sector support workers. We believe the contribution of this participant group is of great value given the repeated contact of older populations and people who self-harm with these groups [13], [14]. Additionally, this study was enriched by a multidisciplinary team, including researchers with diverse areas of expertise and backgrounds, alongside the PPIE group.

Through the use of qualitative methods, our study provides diverse narratives of different participant groups to further understand experiences of self-harm in older adults. Findings from this study highlight the importance of careful consideration when conceptualising functions of self-harm. Adding to previous research in older populations [3], this study offers alternative pathways to understanding the functions of self-harm for older adults, which are more complex than a direct relationship between self-harm and suicidal intent. However, further research is needed to adequately describe the relationships and functions of self-harm in older people, particularly those belonging to minorities, such as ethnic minorities, LGBT groups and those in diverse generational cohorts.

Our study aimed to explore self-harm behaviour in older adults. As further seen in Table 1, only one of the older adult participants (F5) started to engage in self-harm behaviour as an older adult. The experiences of other older participants with self-harm behaviour varied from starting such behaviour in earlier years to mid-adulthood. Future research could explore the potential differences in self-harm in older adults, depending on the starting age of such behaviour to see whether there are differences in such experiences and functions of self-harm in later life.

Current national and international clinical guidance highlight the importance of identifying and supporting older adults who self-harm due to the increased risk of repetition and suicide [1], [10]. Our findings may suggest that self-harm in older adults can be further concealed, which may result in yet lower levels of medical help-seeking behaviour. Clinicians should be aware of the existence of this behaviour in later life, particularly those with mental and/or physical health conditions. It is important that care is taken to avoid deepening any sense of shame or stigma. For this, awareness is needed so attitudes in society towards self-harm change, with wider members of the public, but in particular clinicians and support workers in contact with older populations refrain from expressing negative views towards those engaging in self-harm. Finally, whilst self-harm holds no clear singular function for older adults, it is essential that clinicians supporting older adults, seek understanding individual motivations for self-harm, so the appropriate support and/or referrals are put in place.

Funding

IT is funded by a Keele University ACORN studentship. Additional funding has been granted to IT by Santander Bank and the Allan and Nesta Ferguson Charitable Trust.

CCG is partly funded by West Midlands Collaboration for Leadership in Applied Health Research and Care.

The funders of the study had no role in study design, data collection, analysis, or interpretation. The corresponding author had full access to all the study data and final responsibility for publication submission.

Authors' contribution

IT conceptualised the idea for the manuscript with guidance received from CCG, LD, BB and OB. IT drafted the manuscript and CCG, BB, LD and OB provided feedback. All authors have read and approved the final manuscript.

Declaration of Competing Interest

We declare no competing interests.