A qualitative meta-synthesis of service users’ and carers’ experiences of assessment and involuntary hospital admissions under mental health legislations: a five-year update

Protocol and registration

We prospectively registered the study protocol in the publicly accessible PROSPERO database on 30^th May 2023 (CRD42023423439). The protocol differed from our previous reviews (24, 25) by including service users under 18 and their family carers. We adhered to the updated PRISMA reporting principles (26) and included the completed PRISMA Checklist 2020 in Supplementary Material.

Eligibility Criteria

Inclusion criteria were peer-reviewed qualitative studies published in any language from 1^st January 2018. We excluded books, systematic reviews, commentaries, editorials or grey literature such as pre-prints, dissertations, PhD theses, government reports, conference abstracts. We included studies that collected data through interviews or focus groups as well as auto-ethnographic or case studies. Studies that collected data using questionnaires or surveys were excluded. No restrictions were placed on language or country of study.

There were no age limits for participants. We included papers with service users who had been assessed for compulsory admission or legally detained to a psychiatric hospital and unpaid carers who provided practical and emotional support to such service users. Paid carers and other professionals were excluded.

Key outcomes for which data were sought were service users’ or informal carers’ experiences of either (a) assessment for compulsory admission to a psychiatric hospital, (b) legal detention in a psychiatric hospital and/or (c) appeal and tribunal processes related to detention in a psychiatric hospital. We excluded studies reporting experiences of voluntary and involuntary admission which did not separate these conditions in analyses, and studies which only reported involuntary outpatient treatment in the community.

Data sources

We searched five electronic databases (MEDLINE, PsycINFO, HMIC and Embase via the Ovid platform, and the Social Sciences Citation Index via the Web of Science platform) for articles published between 1st January 2018 and 1st March 2023 (see the search strategy for each database in Supplementary Materials). In addition, we screened the reference lists of all eligible papers for other relevant studies and carried out a forward citation search on all included papers and our previous reviews (24, 25) using Web of Science. Manual screening was completed on 6th July 2023.

Selection process

The resulting list of records was deduplicated in EndNote software before being imported into Rayyan (27) for independent screening. We piloted the inclusion and exclusion criteria on 10 titles and abstracts. All others were then screened independently by one reviewer. A second reviewer double-screened some of each reviewers’ allocation of titles and abstracts, 10% in total. Any discrepancies were resolved between the reviewers. Potentially eligible, full articles were subsequently reviewed independently by two reviewers. Uncertainties were resolved by discussion and in some cases by consultation with a third reviewer.

Data extraction

We adapted the data extraction table developed in the previous reviews for use in the current review. Data extraction was piloted and revised based on 10% included papers using Microsoft Excel. Data extraction was carried out by one member of the research team and checked for accuracy by another researcher. Extracted information included study authors, year of publication, study focus (service user, carer, or both), study setting (country and whether single-site or multi-site), total sample, gender and/or sex, age-range, ethnicity, diagnosis, method of data collection, and method of analysis. We summarised main themes identified by the authors of papers at this stage, prior to meta-synthesis of papers.

Risk of bias and quality appraisal

We used the Critical Appraisal Skills Programme (28) qualitative checklist to appraise the quality of included studies. This included an appraisal of papers along several domains covering research ethics, suitability of methods, recruitment and data collection processes, data analysis, reflection on sources of bias, clarity and validity of findings. Two reviewers conducted quality appraisal independently and discrepancies were resolved through discussion.

Assessment of Confidence in Findings

Two reviewers independently assessed certainty of evidence for each individual finding, defined as the sub-themes, using an adapted GRADE Confidence in the Evidence from Reviews of Qualitative research approach (GRADE-CERQual, (29)). We operationalised components of the GRADE-CERQual to the context of the review as described in Supplementary Materials.

Data analysis and synthesis

We analysed findings (‘Results’ or ‘Findings’ sections extracted verbatim) from included studies using a two-stage process, informed by Thematic Synthesis (30). In stage one, we developed a separate service user and carer initial coding frame to reflect the over-arching themes developed through thematic synthesis and reported in the two previous reviews (24, 25). Following discussion and agreement by the team, we defined and added to the coding frames sub-themes implicit in the published reviews. We tested and refined the two coding frameworks and coding process by double-coding and reviewing two service user and two carer papers. Thereafter, we adopted a deductive approach to code findings from all included studies to the overarching themes. In stage two, the wider project team reviewed relevant data which could not be satisfactorily coded to existing themes and developed additional themes inductively and iteratively. We then modified the initial framework to accommodate newly identified themes and sub-themes, where necessary. All coding was carried out in NVivo 14 (31).

We planned to analyse and report findings regarding service users’ and carers’ experiences’ separately. No formal comparison of sub-groups was planned, but we identified and specifically considered findings regarding groups of people and stages of the detention process for which evidence was limited in our previous reviews (24, 25). These included: young people, people from minority ethnic communities, and the process of assessment for compulsory admission.

Reflexivity

Our research team included people of different backgrounds with prior interest and experience in the topic of investigation. It included senior and junior researchers, people with experience of delivering and receiving mental health care, and people with different ethnic and cultural backgrounds. The team included authors with practitioner experience (psychiatry, nursing and social work), and authors with personal experience of undergoing compulsory detention, or caring for someone undergoing compulsory detention. The team met regularly to plan the study, review study screening and inclusion decisions, contribute to analysis and decisions about themes, and to comment on drafts of the paper – and thus utilised the variety of perspectives to inform the study.

Search results

We report the number of studies identified, screened, and included/excluded through the database search and citation searches in the PRISMA Flow Diagram in Figure 1. A total of 24 eligible papers were identified. Sixteen of these reported experiences of service users only, three of them of carers only. There were five additional articles that contained data regarding both groups. We report study characteristics and our thematic synthesis separately below: first from papers reporting service users’, then carers’ experiences. Papers which reported on both groups are included in both tables, with their relevant characteristics (e.g. number of service user, or carer participants) indicated where appropriate. The quality appraisal of papers is reported in Table 1. Twenty-three studies were rated as high quality, i.e. scored 7 or more on the CASP tool, and one study was rated as medium quality, scoring 4 or above, but below 7.

• Download figure
• Open in new tab

Figure 1.

PRISMA Flow Diagram

Service users

Carers

Certainty of evidence

Certainty of evidence has been reviewed and appraised for each sub-theme, reported separately for service user and carer findings in Tables 4. and 5. respectively. There were 28 service user sub-themes, in which we rated overall confidence as high (10), moderate (13), low (3), or very low (2). For the 19 carer sub-themes, we rated overall confidence as high (13), moderate (4), or low (2). Decisions for lowering confidences were most commonly due to minor or moderate concerns due to relevancy of evidence, coherence of finding, and/or adequacy of data, as documented in Tables 4. and 5.

Main findings

In our two earlier reviews (24, 25) service users primarily reflected on inpatient experiences, whilst data from carers described struggles to find health service support prior to detention and following discharge. In the current update we found a greater overlap in content: both service user and carer data included reflections on experiences of pathways to admission, the extent of therapeutic benefit from inpatient treatment, the frequent experience of coercion and its consequences for future engagement with health services. Studies included in this update also report on recent developments in mental health practice such as crisis planning, advanced directives and street triage, and changes to mental health legislation.

Our findings suggest that the experience of involuntary treatment and compulsory admissions is an often predominantly negative, at times traumatic experience for service users and carers, not always achieving the expected therapeutic benefit. A variety of factors are reported to contribute to this, including the use of coercive practices, too much focus on pharmaceutical treatments, lack of access to psychological and other therapies, uncaring staff attitudes, or a lack of a calm, therapeutic ward environment. Compulsory admissions are often associated with experiences of coercion, and variable but often low levels of involvement in decision making processes, which can both affect the perceived effect of hospital treatment (13). Involuntary treatment may be a particular source of tension as staff are providing care to which service users may not have the capacity to consent or adamantly oppose. Thus, delivering acute mental health services in a collaborative, non-coercive manner in this environment may be a particular challenge.

Our findings indicate that reliance on the use of coercive methods can be therapeutically counterproductive. To minimise experiences of coercion, service users may be motivated to minimise their symptoms and difficulties in discussion with health professionals and family members. Experiences of coercion may in turn make it less likely for people to seek or engage with subsequent treatment, corresponding with other findings (e.g. (1)). Bad experiences of detention and coercive treatment impacts on carers too. If service users mask their symptoms, stay guarded and ‘keep their distance’ as a way of avoiding further bad experiences, it is more difficult for carers to get help for them when needed. In addition to an effect on therapeutic relationships, coercion may also affect family dynamics. Carers described a need for support for themselves at the moment of admission and over time as they assist their family member in managing their mental health for years.

Our synthesis also highlighted some positive experiences during compulsory treatment, for example feeling relief, receiving help to manage symptoms, or family members accessing support. The finding that service users’ views on therapeutic benefit were mixed mirrors previous findings with service users’ groups of varying diagnoses (e.g. (59, 60, 61). In our review, service users reflected positively on being listened to, experiencing flexibility, being offered options. At the same time, as in previous reports, the opposite was also commonly experienced: not being heard, and lacking control over treatment (62). Studies in our review highlighted how staff attitudes (e.g. being respectful, caring, communicating well, empowering) could have an impact on how people saw their treatment and the degree of coercion.

Service user and carer accounts also reflected on the importance of pathways leading to hospital admissions. Several instances were highlighted when either a lack of alternatives to inpatient services, or professionals’ lack of mental health awareness may have escalated a crisis and contributed to involuntary admissions. In contrast, efficient collaboration between services was also noted by participants as a positive experience. An example of this is increasing coordination between mental health and law enforcement services when responding to crises, which is seen as potentially reducing the use of place of safety and police custody facilities in the UK (63).

Some service user reports referred to experiences of racial discrimination, and other inequalities (e.g. due to age, dual diagnosis etc.) in accessing services. This finding is more explicit in our review of recent studies than in previous reviews (24, 25), although it is notable that only a few papers addressed or specifically investigated the experiences of people with marginalised ethnic backgrounds. We note that initiatives to reduce racial discrimination and imparity are seen as increasingly important (10, 46), and should be reflected in additional resources for research and changes to practice in this area.

Whilst our themes and subthemes reported here are focused on the experiences of formal compulsory admissions, we also noted some of the more recent reports (32, 45, 50) described instances when service users underwent treatment whilst technically voluntary service users, but due to fear of coercion and thus de facto detained.

Implications for policy and practice

Although compulsory admissions are an inherently coercive and often aversive situation, service user and carer accounts suggest that there are ways of improving experience and therapeutic potential, and address other issues such as inequality and discrimination:

1. Power, agency and choice were highlighted as important aspects of compulsory admission experience. Increased focus on supporting decision making may be particularly effective, given that many involuntarily admitted service users may have capacity to consent (64), for example by initiatives to provide supported involvement in treatment decisions from early stages of compulsory admissions (65). Additionally, for people experiencing readmission, advance directives and similar tools may support discussion of previous treatment experiences, accessing early support, or reducing the use of coercive practices (46, 66). Evidence shows that care planning and advance statements are effective in reducing detentions (67, 68). Broader use of these initiatives could address this aspect of service user experiences, whilst research is also needed to understand their implementation, and how they work for marginalised groups who may need them most. Providing information in a timely manner during all key stages to the admission process can also have a positive, empowering effect on both service users and carers (2).

2. Service user and carer voices highlighted the importance of working with respectful, engaged, kind staff. This corresponds with earlier qualitative work identifying service users’ desired qualities in mental health staff (69), emphasising the importance of empathy, kindness, respect, effective communication and fostering hope. These qualities have also been identified as facilitators of meaningful service user and staff relationships, and thus supporting meaningful therapeutic engagement (70). Addressing staff shortages, and recruitment and retention of staff with suitable personal characteristics seems important, as it may ameliorate otherwise often negative treatment experiences in the context of compulsory admissions and frequent use of coercive means. Training for mental health and other emergency staff in de-escalation (10), or increasing primary care practitioners’ mental health awareness (71) may also improve access and the quality of initial contact with treatment providers.

3. Service users and carers both identified missed opportunities to avoid detentions. Community-based alternatives to admission such as crisis houses, crisis resolution teams and acute day services may be beneficial both for the individual and the wider community, and wider implementation may have potential to reduce involuntary admissions (5, 12). Both the quality and intensity of upstream community care and availability of community crisis alternatives have been linked to compulsory admission rates (7), highlighting the need for proactive, well-resourced community care. Initiatives to enhance both crisis planning and ongoing monitoring and support post-discharge are of high interest as a potentially effective means to prevent repeat detentions (72, 73).

4. Addressing any inequalities of access to services and reducing racial discrimination and disparities appears to be a key issue to address across several mental health systems (21, 22, 74, 75). Service users from minoritised ethnic groups reported experiences of racism and unfair treatment, thus efforts to promote equality, diversity and inclusion and anti-racist practice are needed, a recent example of which is the Patient and Carer Race Equality Framework in England and Wales (76).

Implications for research

The identified literature reflected accounts and views of some, but not all groups of service users who undergo compulsory admissions. In addition to the low number of papers on experiences of people from ethnic minority backgrounds and young people, more research is needed to understand the experiences of other seldom-heard groups e.g. those with intellectual disabilities, autism, speech impairments, and/or in long-term inpatient care.

With the increase of community crisis service models in recent years (77, 78) we need more evaluation studies to determine which might be most effective in reducing admissions and detention. This could reduce the number of instances in which opportunities to avoid compulsory hospital admissions are missed, a possibility that was identified by papers we reviewed. A clear effect of such alternatives on involuntary admissions has, however, yet to be demonstrated (67), thus, refining such models to ensure a clear focus on people whose history or the severity of their difficulties suggests they are at risk of compulsory admissions is a potentially fruitful direction for service development and research.

Many of the expressed wishes of service users and carers summarised in this review – for clear communication and a chance to be heard, for fairness and respect and some choice and control – map on to the components of procedural justice (79). Procedural justice focuses on perceptions of perceived control and fairness in the process, rather than the outcomes of a situation. It is theorised as comprising four components: allowing citizens a voice; perceived neutrality in decision-making; demonstrating dignity and respect during interactions; and having trustworthy motives. In mental health inpatient contexts, perceived procedural justice is associated with less perceived coercion (80) and better therapeutic alliance with staff (81). Development and evaluation of interventions specifically designed to enhance procedural justice in the assessment and compulsory detention processes are of high interest. Recent examples of this include crisis planning and monitoring for detained patients initiated in hospital (72), or supporting involvement in treatment decisions from early stages of compulsory admissions (65).

Strengths and limitations

We used robust systematic search methodologies and followed established guidelines both for newly published primary research and for synthesising results. We have included recent studies that allowed corroborating and refining findings from previous systematic reviews and identify relevant new themes. Our review also reported findings from service user and carer groups that were present less in previous works, for example people from marginalised ethnic backgrounds. Our team had a diverse set of skills to draw on including researchers with lived experience, which informed the depth of our analysis, interpretation, and writing of the manuscript.

The use of qualitative synthesis methods may have led to the loss of more nuanced information on service user and carer experiences available in individual primary studies. There were only a few new studies available on the compulsory admission experience of ethnic minorities, or young people. Similarly, only three papers focused specifically on carers’ experiences, suggesting a necessity of further research in these areas. Our review focused on the experience of compulsory admission under mental health legislation, thus it did not fully explore involuntary treatment experiences when formally voluntary service users encounter coercive practices and undergo forced treatment despite their status. We also acknowledge that lived experience involvement could have been heightened by involvement in the earlier stages of the project, for example when drafting the review protocol. Evaluation of certainty of evidence for pre-existing themes has been carried out on newer information only, as the use of GRADE-CERQual was not a PRISMA recommendation at the time of the original review.

by Karen Machin, Patrick Nyikavaranda and Tamar Jeynes

This revised review builds upon two previous reviews by the same team from 2018. These earlier reviews, enriched by lived experiences commentaries, identified a pressing need for enhancing patient and carer experiences. They also highlighted the sluggish pace of translating knowledge into practical action. Regrettably, this latest review offers minimal new insights. However, it does underscore the counterproductive nature of over-reliance on coercive methods in care. The review acknowledges that while some may find solace in any support, it’s deeply troubling that issues like racial discrimination, unequal access, and general dissatisfaction have become more pronounced in recent years.

Notably, research by Morris et al. (82) suggests that it can take up to 17 years for research findings to be implemented in clinical practice. We cannot continue to wait. Urgent action is needed to enhance the experience of detention and, or even, dare we say it, earlier supportive interventions. This could address the immediate needs of individuals and potentially reduce the frequency of detentions.

As authors, we might argue that one way to promote change might be for researchers, funders, policymakers, and practitioners to finally act on what people with lived experience tell them they need. The current review suggests several improvements, including the provision of community-based alternatives to detention and the availability of family support. Additionally, it highlights the critical need for genuine cultural awareness training, moving beyond mere procedural compliance.

The numerous television exposes and media reports depicting the harrowing experiences within inpatient wards have understandably deterred many from seeking voluntary support. It is disheartening that this systematic review merely reiterates what service users and carers have long known: significant change is imperative. Perhaps the time has come for service users and carer groups to be more active in setting research agendas and conducting research themselves.