Introduction
Romantic relationships are a ubiquitous part of human life and can be defined in a variety of ways. Broadly speaking, romantic relationships may be distinguished from platonic friendships due to the presence of sexual or increased physical intimacy. However, there may also be differences in terms of level of care shown to a romantic partner, romance, love and exclusivity [
1]. Qualitative findings suggest people who experience psychosis associate having a partner with recovery [
2,
3], however, the intimate relationship needs of this population are often unfulfilled. People with mental health difficulties experience higher rates of relationship breakdown than the general population [
4]. Additionally, a meta-analysis of 1404 participants with a schizophrenia diagnosis and mean age of 39.9 years old, found just 15.6% were married [
5] – considerably lower than the national average [
6]. Furthermore, participants were significantly less satisfied with their sexual relationships than any other life domain.
One factor related to dissatisfaction with intimate relationships is the impact of side-effects from antipsychotic medications, such as sexual dysfunction [
7]. However, prior to this, people with psychosis face numerous barriers to forming the romantic connections that lead to sexual relationships. For example, experiences of discrimination [
8] and the internalisation of stigma which may make individuals feel undesirable [
9]. A qualitative study which recruited people with a diagnosis of psychosis found low self-esteem and experiences such as hallucinations or becoming withdrawn were seen as barriers to both forming and maintaining connections with partners. Finally, participants discussed how sexual abuse (experienced by 36% of the 28 participants) negatively impacted self-worth and made it difficult to trust others or enjoy physical intimacy [
10]. More recently another study interviewing women with ‘serious mental illness’, concluded participants expected to be rejected by those without mental health difficulties. Women minimised the importance of having a partner due to the disproportionate caretaking burden placed on them in heterosexual relationships and a desire to prioritise their own mental health. In line with previous studies, sexual trauma reduced the ability to enjoy sexual intimacy. Additionally, ‘gaslighting’ partners dismissed women’s genuine complaints within relationships as symptoms of their mental health difficulties [
11]. As such, recommendations have been made for research to further investigate the intimacy needs of people with psychosis, with a view to developing trauma sensitive interventions that address stigma and enhance satisfaction in this area of people’s lives [
4,
11‐
13].
Currently, interventions regarding the intimate relationships of people with psychosis and other mental health diagnoses tend to focus on sexual health rather than the formation and maintenance of healthy romantic relationships. There are few known psychosocial relationship interventions where sexual risk reduction is
not the primary outcome [
14‐
16] and no known published studies within the UK that have delivered a psychosocial intervention around romantic relationship issues to people with experience of psychosis. Additionally, and of concern, there are no known studies that have sought to understand what support, if any, people who experience psychosis want from mental health services regarding intimate relationships.
All authors of this paper view romantic relationships as a fundamental part of human life that mental health services often neglect [
17]. This is something I (RW) observed whilst working in mental health services. It inspired the current research. If services are to provide support to address this area of unmet need then it is important that the delivery and content of any intervention reflects the self-defined requirements of the recipients. As such, this study aims to:
-
Investigate how people who experience psychosis conceptualise romantic relationships
-
Identify whether/how people with experience of psychosis would like community mental health services to provide support with romantic relationship issues
Given the nature of our research questions, our epistemological approach and wanting to gain a deep understanding of participants’ views, a qualitative methodology was adopted. Semi-structured interviews were chosen as they are ideal for understanding the views of participants and exploring sensitive topics [
18].
Methods
Study design
Semi-structured, one-to-one interviews were conducted. We considered this study to be experiential qualitative research, as the aim of the research questions was to understand the views of people with experience of psychosis and to prioritise their voices [
18]. Data were collected and analysed from a critical realist social constructionism perspective which, while acknowledging that knowledge is always mediated through social processes, suggests that discourse indirectly reflects an underlying reality. In other words, such an approach is “… ontologically realist but epistemologically relativist” [
19 p. 92]. In addition, we saw our data as being co-constructed and influenced by the interaction between interviewer and participant [
18].
RW is a white British, middle-class, heterosexual, 31 year old, female PhD student with a long-term partner. She has previously worked as a support worker in community and inpatient mental health services. MH identifies as a white Canadian middle-class, heterosexual, middle-aged, married woman with over 20 years of experience in the mental health field as a clinician and researcher. Her PhD in clinical psychology focused on gender and psychosis. GH and FV are qualified clinical psychologists, experienced in both conducting research, and working with people who experience psychosis clinically.
Participants
Participants were eligible if they: 1) had either a diagnosis of psychosis or experience of psychosis that met the criteria for acceptance into early intervention services, 2) were currently receiving support from community-based mental healthcare services and 3) were aged 16 years or over. Convenience sampling was used initially, followed by purposive sampling to answer our research questions and get a deeper understanding of participants’ experiences. We considered that social identities including gender, sexuality, and ethnicity may influence people’s views and experiences and attempted to recruit a diverse sample with an even representation of men and women as well as both single and partnered participants. Participants were recruited via two community mental health services in the North West of England and through other research studies being conducted at the University of Manchester, where individuals had given consent to be contacted. Ten participants (six male, four female) were recruited. Participants’ ages ranged from 21 to 64 years (M = 29.7 for men, 50.0 for women). Seven were receiving support from Early Intervention Services, Recovery/Community Mental Health Teams, three were solely under the care of a psychiatrist. All had previously experienced a romantic relationship but, at the time of interview, four were in a long-term relationships. Of these, three were living with their partner. None were currently employed, three were attending educational courses. Nine participants identified as heterosexual and one as gay. Seven identified as white British/English/Irish. Two participants identified as white British with mixed heritage (Polish and Afro-American) and one as French Jewish.
Data collection procedure
This research was carried out in accordance with the Declaration of Helsinki and ethical approval was given by Greater Manchester South Research Ethics Committee (18/NW/0755). A mutually convenient time and place was arranged to conduct the interviews. Four interviews were conducted at the participants’ home address, four on NHS sites, one at the University of Manchester and one over the telephone. All interviews were conducted by RW. A topic guide was developed by RW and MH consistent with the thematic analysis approach and wider literature on qualitative interview schedules [
18,
20,
21]. Questions on the topic guide included, amongst others: ‘How do you think friendships are similar/different to the relationship you might have with a romantic partner?’, ‘Could you describe a romantic relationship that you have had, what was it like?’ and ‘Have/would you ever ask(ed) your care co-ordinator for support with a romantic relationship issue? Why/why not?’. The topic guide was revised after several interviews to include questions attentive to abuse, since this issue was raised by several participants. Most interviews were completed in one session, four were conducted over two sessions. Interviews were audio-recorded and reflective notes were written after each interview. Some participants reported initially being apprehensive about the sort of questions they might be asked but felt less anxious once the interviews were underway. The mean length of interviews was 57 min for men and 1 h 51 min for women. The lengthier interviews with women may have been due to RW being more easily able to develop rapport with female participants, being female herself and/or, becoming more skilled as an interviewer during the process of data collection.
Data analysis procedure
Interviews were transcribed verbatim and anonymised by either RW or research interns. Braun and Clarke’s reflexive thematic analysis was adopted to identify common patterns across transcripts [
18]. Transcripts were read several times before all data relating to the research questions were coded. Codes and themes were developed inductively. A hybrid approach to coding was adopted whereby some codes generated were semantic (reflecting the explicit/surface meaning of the data) and others were latent, meaning they moved beyond the literal meaning of the data to include implicit theoretical ideas such as stigma, a prominent theme in interviews [
18,
22]. Once ten interviews had been coded, candidate themes were developed and data which opposed these themes were highlighted. Candidate themes were discussed with MH, then reviewed and revised by RW and discussed with the whole research team. The analysis was subject to several iterations of review and revision. When nearing completion, RW attempted to contact nine participants who had given consent for member checking. A summary of the analysis was sent to six who were contactable/agreed to take part in the process, four (two male, two female) provided feedback. The purpose was to help ensure participants’ opinions had been accurately represented. Where participants disagreed with the analysis, the goal was not to reach agreement but to try to understand the reasons for differing views [
23]. Feedback was largely in agreement. Sections regarding fear of rejection by partners and the power imbalance between mental health professionals and service users particularly resonated. Some gave alternative views around parenting, caregiving in relationships, and were less critical of mental health services - these views have been incorporated into the analysis. Quotes included in the results were ‘cleaned’ by removing repeated words and non-verbal utterances to improve readability. Ellipses were used to indicate words before and/or after a quote that were not included and ellipses in square brackets were used to denote where sections of data were cut.
Discussion
This study, the first to our knowledge, aimed to investigate how people with experience of psychosis conceptualise romantic relationships and explore if, and how, they would like to be supported by mental health services in this area of their lives. Romantic relationships were conceptualised as a fundamental aspect of life. While sex was seen as a distinguishing feature, in line with the literature on intimacy, emotional and cognitive intimacy were also viewed as a central and desirable dimensions of romantic relationships [
25].
Themes of prejudice, discrimination, and rejection due to experience/diagnosis of psychosis were prominent and cross-cutting in our findings. Indeed, reductionist terms such as ‘schizophrenic’ are still common and appear alongside graphic and emotive descriptions of violence in the media, contributing to persistent beliefs within the general population that people who experience psychosis are inherently dangerous [
26]. A previous study found people with a diagnosis of schizophrenia felt a need to conceal their diagnosis from friends and family [
27]. The present study found this was applicable within romantic relationships also. A novel finding is how some navigated this predicament – using partial disclosures or enquiring about a partner’s attitude towards mental health tangentially to gauge reaction whilst protecting themselves from judgement and rejection. Our findings are also consistent with previous research which has linked experienced stigma with internalised stigma and low self-esteem [
9]. Although some participants in the current study rejected stereotypes, medicalised language was still used to discuss mental health, suggesting an illness model of psychosis had been internalised. Belief in biological rather than social causes of psychosis has been linked to increased negative attitudes towards people with psychosis diagnoses [
28,
29] and thus may have contributed to feelings of undesirability among participants. It is important mental health services are aware that people who experience psychosis may expect rejection from romantic partners due to internalised stigma, discrimination and lack of self-worth [
8,
9]. Consistent with the literature, parenting was another area where the impact of stigma was apparent in participant accounts. An Australian study found many parents with a mental health diagnosis felt others believed they were incapable of being a good parent due to their mental health. Although the majority rejected this idea and reported being a parent was a motivating factor to better manage their mental health, this perceived stigma was still associated with negative self-reported parenting experiences, with mothers more likely than fathers to internalise stigma [
30]. Subsequently, the authors recommended interventions to target stigma [
30]. However, their finding that participants had also experienced discrimination from services, and our key sub-theme regarding power imbalance leading to mistrust of services, highlights the importance of any such interventions being wanted, and delivered in the context of a strong therapeutic alliance that empowers recipients.
Participants in this study largely rejected the idea of mental health services providing support regarding romantic relationships, however this is not to say mental health services should abandon any idea of providing support in this area of people’s lives, or that some service users would not benefit from support, particularly regarding internalised stigma and self-esteem. Rather, rejection of support was mainly due to power imbalances and a lack of an effective therapeutic alliance. Interestingly, in our previous qualitative study, mental health professionals identified age and gender as important influences on relationship dynamics [
17]. However, participants in this study focused on relational aspects such as time spent during visits/appointments, and criticised professionals’ emphasis on symptomatology over their personal lives. The power imbalance between mental health professional and service user experienced by participants had also damaged trust. Trust enables the management of anxiety in times of vulnerability, but when we suspect others are not acting in our best interests trust in undermined [
31]. In mental health services trust can be compromised by a lack of time for relationship-building between service user and professional as well as involuntary treatment [
31]. Our previous study found mental health professionals may be concerned about the risk to service user wellbeing and their career if supporting someone with a romantic relationship issue had a negative outcome [
17]. Managing these conflicting interests and developing trust is essential for a therapeutic alliance, which our findings indicate is a prerequisite for any support/work around romantic relationships. Empowering service users regarding decision making has also been linked to trust [
32]. In line with the literature, participants in this study valued and wished to protect their autonomy in any discussions or decisions about romantic relationships. Mistrust and concerns they would not be given this power appeared to be the primary reason participants did not want support from mental health services in this area of their lives. A recent ethnographic study within an early intervention team concluded when managers engage in reflective leadership and resist organisational policies seen as counterproductive with the goals and values of the service, workplace trust can develop between staff i.e., supportive relationships and shared decision making where possible. This facilitated knowledge sharing and collaboration. Staff were able to take risks trying new approaches which aimed to and succeeded in, generating trust between the service and clients [
33]. Just as trauma-informed services need to provide trauma-informed training and supervision to staff [
34], mental health professionals need to be able to trust in their service to support them in order to facilitate trust between themselves and service users. Regarding support with romantic relationships specifically, mental health professionals need to have trust in management support and their career security to feel confident doing work in this area that is empowering and does not compromise service users’ autonomy.
The only area that participants did agree mental health services should provide support in was instances of abusive relationships. The prevalence of interpersonal abuse in people who experience psychosis and schizophrenia is known to be high [
35,
36], but less is known about intimate partner violence specifically. A systematic review and meta-analysis which aimed to understand the prevalence of domestic violence in people with mental health diagnoses found only three studies measured domestic violence in people with schizophrenia/psychosis. However, these reported a prevalence ranging from 43.8–83.38%, with one finding women with psychosis were more than three times as likely to report violence from a partner in the past year compared to those without a mental health diagnosis [
37]. Intimate justice theory suggests stigma, inequity and discrimination can reduce an individual’s beliefs about what they deserve and are entitled to within a relationship [
38]. By extension, marginalised groups, such as those receiving a psychosis diagnosis, might be more likely to have lowered expectations of partners due in part to internalised stigma and dominant negative discourses surrounding psychosis. This is similar to our findings on expectations of rejection and warrants further research. Unfortunately, a recent meta-review concluded there are few interventions which specifically address intimate partner violence in women who have a mental health diagnosis [
39]. Based on this finding, Van Deinse and colleagues called for professionals working in both mental health services and domestic violence/sexual assault services to receive training in delivering interventions that meet the intersectional needs of this group [
39]. Although less researched, men also report abuse from romantic partners [
40]. Male, in addition to female participants in our study, disclosed intimate partner violence, highlighting both men and women with experience of psychosis may benefit from support in this area. Some participants endorsed the idea of peer support. Peer support may be a helpful approach to reducing internalised stigma due to increasing social connectedness, providing role models with shared experience and removing the hierarchical power dynamic of mental health services [
41].
Limitations
This study recruited a relatively diverse sample regarding age and included those who were single, partnered, as well as those with and without children. However, we were not able to recruit a more representative sample of social identities such as ethnicity and sexuality. The findings present a somewhat broad view of the beliefs people with experience of psychosis have on the topic and may be transferable to similar participants receiving support from community mental health services in the UK. However, future research may benefit from recruiting more homogenous samples, to better understand any specific needs of these groups. Additionally, further attention should be paid to whether and how social identities such as age, gender, ethnicity and sexuality influence how people wish to be supported in this area of their lives.
Conclusions
Findings from this study indicate that people with experience of psychosis view romantic relationships as a vital part of life. However, mental health services were often seen as being uninterested in the romantic relationships of participants and more interested in symptoms. This juxtaposition in attitudes contributes to the dehumanisation of mental health service users and is damaging to the therapeutic alliance. Simple ways professionals can overcome this and foster more egalitarian therapeutic relationships were highlighted, for example, by showing interest and engaging in conversation about the significant others in service users’ lives .
Stigma and discrimination regarding people with experience of psychosis were prominent in participants’ narratives and were seen as barriers to establishing romantic connections. Use of psychosocial explanatory frameworks may help overcome the pervasive stigma and discrimination around psychosis and promote more compassionate and positive attitudes [
29]. Whilst it is clear many participants in this study would not currently welcome the involvement of mental health services in this area of their lives, there was agreement support should be provided to help those in abusive relationships, highlighting the importance of routine enquiry and services being trauma-informed. Future research and practice could draw from intersectional approaches to working with intimate partner violence [
42] to address this identified need [
41].
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.