Experiences of in-patient mental health services: systematic review

Abstract

Background

In-patients in crisis report poor experiences of mental healthcare not conducive to recovery. Concerns include coercion by staff, fear of assault from other patients, lack of therapeutic opportunities and limited support. There is little high-quality evidence on what is important to patients to inform recovery-focused care.

Aims

To conduct a systematic review of published literature, identifying key themes for improving experiences of in-patient mental healthcare.

Method

A systematic search of online databases (MEDLINE, PsycINFO and CINAHL) for primary research published between January 2000 and January 2016. All study designs from all countries were eligible. A qualitative analysis was undertaken and study quality was appraised. A patient and public reference group contributed to the review.

Results

Studies (72) from 16 countries found four dimensions were consistently related to significantly influencing in-patients' experiences of crisis and recovery-focused care: the importance of high-quality relationships; averting negative experiences of coercion; a healthy, safe and enabling physical and social environment; and authentic experiences of patient-centred care. Critical elements for patients were trust, respect, safe wards, information and explanation about clinical decisions, therapeutic activities, and family inclusion in care.

Conclusions

A number of experiences hinder recovery-focused care and must be addressed with the involvement of staff to provide high-quality in-patient services. Future evaluations of service quality and development of practice guidance should embed these four dimensions.

Declaration of interest

K.B. is editor of British Journal of Psychiatry and leads a national programme (Synergi Collaborative Centre) on patient experiences driving change in services and inequalities.

Patient experience is a vital source of evidence that can drive the provision of high-quality health services.^{1, 2} Mental health in-patients report a range of experiences including fear of assault, concerns regarding coercion, limited recovery-focused support and lack of therapeutic activities.^3–8 A triennial review of mental health services in England by the Care Quality Commission (2017)⁹ highlighted several serious concerns about in-patient care, including wards located in older buildings not designed to meet the needs of acute patients, unsafe staffing levels and overly restrictive care in wards far from patients' homes and families.

The National Health Service (NHS) is under pressure to deliver timely, effective and affordable care with increasingly constrained resources. The National Institute for Health and Care Excellence, the NHS National Quality Board and others have restated core principles of patient-centred care including dignity, compassion, choice and autonomy,^{3–5, 5–8} and called for a strengthening of the patient voice. Healthcare providers are now required to collect data to assess patients' experiences of care.^9–12 However, the impact of this data collection on services is unclear¹³ because of: the diverse and poor-quality feedback methods,¹⁴ a lack of consensus about which experiences are most salient (and hence should be asked about), and limited evidence about how patient experience data can guide service improvements.^{13, 15} Such challenges highlight the need for robust evidence to inform best practice, with clarity about the experiences of most importance to patients. In response to this need, this systematic review aimed to identify the most salient experiences of people using in-patient mental healthcare to inform the provision of high-quality services.

Method

The review was divided into a scoping review to ascertain the nature and size of the evidence base, and the main systematic review.

Protocol and registration

The EURIPIDES (Evaluating the Use of Patient Experience Data to Improve the Quality of Inpatient Mental Health Care) systematic review was registered in 2016 on PROSPERO: CRD42016033556.

Scoping review

Before the systematic review, a scoping review was conducted to ascertain the extent, range and nature of studies to map emerging key themes without describing the findings in full or performing a quality check¹⁶ and to inform the main review. Six key authors known to be experts in mental health patient experience were contacted for new or unpublished reports and studies.

Patient and Public Involvement Reference Group

The Patient and Public Involvement Reference Group (PPIRG) included 10 service users, recruited by the Mental Health Foundation, with experience of in-patient care or caring for someone who had been an in-patient. They were invited to two meetings: first, to obtain their views on the themes identified in the scoping review, with the potential to add further concepts they felt had not been identified; and second, to obtain their opinions on themes identified in the main systematic review and to contribute to the interpretation of our findings. A full description of the patient involvement in the study is reported using the GRIPP2 Short Form Checklist in Table 1.

Table 1 Reporting Patient and Public Involvement in the EURIPIDES study using GRIPP2

a. Staniszewska S, Brett J, Simera I, Seers K, Mockford C, Goodlad S, Altman DG, Moher D, Barber R, Denegri S, Entwistle A, Littlejohns P, Morris C, Suleman R, Thomas V, Tysall C. GRIPP2 reporting checklist: tools to improve reporting of patient and public involvement in research. BMJ 2017; 358: j3453.

Identification of studies for the systematic review

Guided by the themes that emerged from the scoping review, search terms and a search strategy were developed and applied to the databases MEDLINE, CINAHL and PsycINFO. An example of search terms and results is reported in Fig. 1. Reference lists of included papers were scanned. The search deviated from the protocol in that only three of five databases were searched due to the large numbers of abstracts retrieved (Web of Science and Embase were not used).

Fig. 1 Example of search strategy from MEDLINE.

Inclusion and exclusion criteria

All study designs were considered if papers included experiences of current or former in-patients of mental health institutions. No restrictions were applied based on country. Articles were included if they reported primary research, were peer reviewed and published in English between January 2000 and January 2016. Papers were excluded if they were not primary studies, based on pre-2000 data, included children and adolescents (aged under 18 years) or were not in the English language. Where study participants included both in- and out-patients, only data regarding in-patient experiences were extracted. Reviews (Table A.1) were noted and reference lists scanned, but excluded from the review to avoid bias.

Study selection

Titles and abstracts were screened (C.M., G.C.), 20% of which were independently cross-checked for agreement before obtaining full-text articles (S.S. and C.M.). Full texts were obtained where the abstract was unclear. Any disagreements could be resolved by consensus (C.M., G.C. and S.S.) but no disagreements occurred.

Data extraction

The data extracted, using Microsoft Excel (version 2013), included citation details, sample recruitment and research methods, findings related to key concepts and any other emerging concepts (C.M.).

Quality and risk of bias in individual studies

The quality of the studies were evaluated by the Critical Appraisal Skills Programme (CASP) Qualitative Checklist,¹⁷ undertaken by C.M. Because of the heterogeneity of the included studies, many of which were descriptive in their approach, this checklist provided an appropriate basis for comparison between studies. The only question change in the CASP checklist was from ‘Is the qualitative methodology appropriate for this study?’ to ‘Is the methodology appropriate for this study?’

Data analysis

The scoping review informed the development of a thematic framework, which guided but did not restrict the review. A narrative synthesis of the themes was undertaken.¹⁸ As the researcher read each study, an initial preliminary synthesis of the study was undertaken and emerging sub-themes were identified. The researcher was then able to compare themes and sub-themes within and across studies and further develop them into the main themes. Themes were summarised in a descriptive form, allowing for the findings of all review studies, regardless of study design, to be aggregated and summarised. We used the concept of data saturation to help us decide when to complete data extraction. Saturation of data is judged to have happened at a point where no new themes are being identified in the studies when compared with what has already been extracted.⁷ It is a useful approach for large reviews where the addition of further papers is unlikely to change key findings.

Results

PPIRG

Key themes identified in the scoping review were discussed in detail by group members who critiqued their content and identified additional areas such as boredom. The PPIRG provided content and face validity for the identified themes and provided real-life examples of the themes from their own experiences. The PPIRG also provided an opportunity to check if the themes identified from international studies resonated in a UK context.

The systematic review

A total of 4979 abstracts were screened and 116 papers fulfilled the inclusion criteria (Fig. 2). Two consecutive sifts were conducted due to an error in the first search of the PsycINFO database omitting 2980 hits which was identified after the first sift was completed. The first sift of 1999 hits resulted in 72 relevant papers for the review; 11 papers were from same studies.^19–29 Following this, the second sift of 2980 abstracts resulted in an additional 44 studies fitting the criteria (total n = 116). Drawing on the principles of data saturation,³⁰ additional studies that repeated themes already identified were excluded from the main review. In total, eight studies added new themes and were included at this stage.

Fig. 2 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2009 flow diagram.

A total of 16 systematic reviews (Table A.1) which investigated in-patient experience were identified. In total, 72 studies were included in the review, a third of which were from the UK^24–47 (n = 24)^{19–21, 25, 27, 31–49} (Supplementary Table 1 available at https://doi.org/10.1192/bjp.2019.22). Although studies using qualitative methods were most common (Table A.1), studies using patient experience questionnaires and patient record data were also included. The CASP checklist identified many of the papers as being of medium to poor quality.

Timing of data collection in included studies

Little information was provided about the timing of data collection in over a third of papers (37%), other than describing participants as in-patients at the time.^{25–27, 31, 32, 35, 36, 43, 44, 48–63} Data were mostly collected just before,^{28, 29, 45, 64–73} immediately after discharge^{20, 45, 59, 74, 75} or from former in-patients.^{22, 23, 34, 37–39, 41, 42, 46, 47, 63, 76–80} This suggests that patients were recovering when experiences were elicited. In three studies, data collection coincided with a ward event (e.g. refurbishment).^81–83 A number of studies (n = 12, 17%) collected data shortly after an event such as admission,^{19, 21, 84–86} seclusion, sedation or restraint.^{24, 33, 87–92}

Identification of key themes

Patient experiences were categorised into four overarching themes or dimensions of experience: the importance of high-quality relationships; averting negative experiences of coercion; a healthy, safe and enabling physical environment and ward milieu; and authentic experiences of patient-centred care. These key themes accompanied by sub-themes are described in detail below.

The importance of high-quality relationships

The importance of high-quality relationships was the most consistently reported theme.

Important factors in developing such relationships with staff included being treated with respect, feelings of stability, recognising empathy and high-quality communication^{19, 23, 24, 27, 28, 35, 36, 38, 39, 51, 55, 60, 61, 63, 78, 87, 90} with staff who patients felt were trustworthy, reliable^{35, 63, 69} or helpful.^{27, 51, 54, 62} Good staff–patient relationships facilitated the in-patient care pathway in mental health institutions^{28, 35, 39, 51, 68} and reduced the use of coercive measures.^{35, 45, 78} Ward rounds were an important setting for staff–patient interaction and patients reported these as helpful and informative.⁴⁴

Potential barriers to therapeutic relationships included: gender-specific problems – male nursing staff were not welcome if the patient had a history of abuse by male perpetrators^{36, 78} or where gender-specific cultural barriers existed (e.g. a Muslim woman supervised by a male nurse);⁶⁸ lack of meaningful communication – where communication was compromised due to differences in culture, language, religion,^{34, 39, 57, 68} through use of coercive measures^{33, 60} or where technical language used by staff was not easily understood;¹⁹ absence of regular ward staff – patients were upset by the absence of regular ward staff due to office duties, shift working, reliance on temporary staff^{23, 24, 27, 28, 35–37, 39, 45, 46, 51, 54, 55, 63, 69} and having extended waits to speak to staff^{24, 36, 46, 54, 77, 80, 82} particularly at ward rounds;⁴³ poor staff attitude – where patients complained that staff ignored them,^{57, 87, 88, 91} displayed indifference²⁴ or insufficient understanding of patients;⁷⁸ inconsistent staff behaviour – reports of staff interpreting ward rules inconsistently, causing confusion;^{19, 23, 27, 31, 33, 36, 46, 49, 82, 91} staff abuse – some patients reported abuse by staff, including provocation, bullying, shouting or belittling of patients.^{19, 23, 27, 28, 33, 39, 56, 62, 78, 79, 83, 87, 88}

Relationships with other patients and with relatives

Patients relied on other patients for information about ward activities and rules, to share experiences and when debriefing after group sessions.^{22, 45, 77, 82, 83} However, arguments and violence between patients^{36, 39, 48} generated fear and isolation for some, causing them to retreat to their rooms for safety or to abscond.^{23, 37, 39, 49, 65, 80}

Isolation from family caused distress. Patients reported that having a friend or family member with them would have helped with orientation⁷⁹ and they could have helped staff with assessments and treatment plans.^{22, 38, 53} However, family members felt left out of decision-making about care.⁹²

Averting negative experiences of coercion

The second main theme was concerned with experiences of coercion. All patients expected to be treated as ‘normal human beings’^{24, 29, 77} and addressed professionally, including during restraint.⁸⁷ Patients wanted the reasons for coercive measures to be communicated so they could understand them as this helped some patients trust staff and feel safe.^{46, 67, 75, 79, 87} Patients valued persuasion over threats of force⁶⁰ and coercion,⁷⁸ which could bring back memories of past history of violence and neglect.^{33, 88, 89}

Where coercive measures were discussed in the studies, these included experiences of sedation, seclusion and restraint. It has been reported that Black and minority ethnic patients are more likely to experience coercion than White patients.

Ethnicity

Two studies examined the commonly held perception that Black and minority ethnic patients experienced more coercion on admission than other patients.^{21, 74} The findings were not conclusive: although hospitals in the UK with higher proportions of Black and minority ethnic patients employed more coercive practices, this was independent of individual patient ethnicity.^{21, 74}

Sedation

Some patients recognised that medication was important for the in-patient care pathway.^{20, 39, 41} Some trusted staff to decide on appropriate sedation,^{32, 52} whereas others felt empowered to decide on timing and dose of medication when administered on an ‘as needed’ basis.³² However, patients also voiced concerns that included lack of communication about consent, information about medication and advanced wishes;^{39, 52} lack of confidentiality regarding medication;^{32, 42} perceived overmedication^{32, 39, 41, 46, 47, 52, 69} (including overlooked or ignored reports of side effects);^{28, 41} and fear of harm during forced medication,^{20, 32, 39, 54, 60, 78} for example patients in crisis reported a fear of being raped by staff or of dying.^{20, 41, 78, 88}

Seclusion

Some patients reported seclusion as helpful or necessary^{24, 57, 79, 88} and that they felt safe as staff were nearby.^{24, 57, 88, 90} Patient concerns included having insufficient information about the reasons for seclusion^{23, 24, 46, 57, 88} before or after the event.^{24, 57} Seclusion was perceived as a punishment⁷⁹ and associated with limited contact;^{57, 88} lack of concern by staff;⁸⁹ degradation and humiliation, e.g. lack of facilities^{24, 57, 89} or being stripped of clothing in front of staff members;^{61, 79, 89, 91} and violation of rights⁸⁸ and dignity.⁶¹

Restraint

Restraint was described as forcible manual or mechanical restraint and typically involved several staff, mostly nurses^{23, 60, 78, 88, 92} but occasionally security staff.^{78, 92} Restraint was described negatively^{25, 33, 78} and fear of restraint prevented patients from seeking help earlier.³³ There was a risk of harm if mechanical restraints were used,⁸⁷ although these were not used in all countries. Talking with staff following restraint or being allowed to examine records of the event was considered helpful.³³

In addition to the use of coercive measures, patients also described perceived punishment by staff^{19, 35, 41, 80, 91} in the form of the removal of leave entitlements,³⁵ removal of furniture and personal items^{41, 91} and not being able to stay up in the evening.^{19, 80} Patients described this as a violation of their rights.^{23, 57, 58, 88}

A healthy, safe and enabling physical environment and ward milieu

The third main theme focused on a healthy, safe and enabling environment. This contributed to how relatives felt when visiting,⁹² how patients felt about themselves³⁹ and how they reacted to treatment.^{36, 39, 42} Johansson et al (2003)⁶³ argued that the physical environment was as important to patients as receiving satisfactory care. A number of studies reported that patients saw hospital as a ‘sanctuary’⁸⁰ or a ‘safe space’⁶² where they could have time to reflect away from day-to-day stressors,^{38, 50} be kept safe^{19, 39, 48, 54} and experience a caring, therapeutic environment.⁸⁰

Patients felt that their in-patient care pathway was aided by connection to the ‘real world’⁶¹ and that being made to feel ‘normal’^{24, 28, 51, 77} was important. This included being allowed to walk around hospital grounds.^{39, 80} Older establishments often had extensive grounds and patients reported that access to these spaces resulted in less need for medication.³² Access to a place of worship was comforting,^{51, 68} as was freedom to make small decisions^{31, 41} such as making snacks⁶² or hot drinks.³⁶ Private bedrooms were important,⁸⁰ being near windows enabled ward-bound patients to enjoy the outside and fresh air,⁸³ and appropriate use of colour was described as conducive to recovery.⁸⁰ An environment where staff and patients mixed together reduced feelings of stigma⁵¹ and encouraged favourable interactions.⁶³

Patients reported several environmental problems that were not conducive to recovery-focused care. Some of these were associated with arguments and violence between patients.^{36, 39, 48} Other environmental problems included noise from doorbells, alarms and telephones.⁸² Poor positioning of the nurses' stations often created physical divisions between patients and staff, reducing interaction.^{61, 80, 92} Communal spaces sometimes lacked privacy for visiting relatives or opportunities for physical activity,⁴⁹ especially for those under close observation.⁹²

There were also contradictory reports. In several studies, some patients described hospital as a place of confinement rather than therapy.^{19, 29, 36, 37, 39, 42, 80} There were analogies with prison^{29, 36, 39, 42, 80} and punishment.^{37, 39} This was particularly so in secure units with a lack of outside space³⁹ and where more patients were admitted compulsorily.²⁹

Ward milieu

Related to environment was the experience of ward milieu, which was shaped by the conduct of staff. Staff provided structure, order and safety⁸² and were responsible for creating a congenial atmosphere.⁵⁴ Feeling safe was a prime concern to patients^{48, 65} who perceived wards to be safe when they viewed staff as trustworthy,³⁵ caring and supportive.^{35, 38} Wards were sometimes criticised as being too busy^{36, 49, 54} and reactive to events such as restraint,^{56, 79, 92} seclusion⁹¹ or violence.^{23, 58, 80} Patients felt vulnerable to the latter,^{23, 37, 39} fearful of other patients^{49, 78} and worried about security of belongings.^{36, 65, 80} Fear contributed to withdrawing within the ward^{49, 81} or leaving hospital.^{37, 80}

Ward routines also shaped patients' experiences. The day⁵¹ was often structured to include individual and group therapies as well as other activities, e.g. puzzles, conversation or listening to music.⁹² Evenings were typically less structured.⁵¹ Some patients relished the leisure time^{24, 38, 50, 54} and some took this as a time for personal reflection.^{38, 51, 57} However, others were uneasy^{38, 51} and reported insufficient^{36, 49} activity. ^{23, 24, 39, 49, 68} The location of the hospital – being close to family – was important to patients⁷⁹ and they appreciated the inclusion of, and support from, families.^{22, 38, 53}

Boredom

‘Boredom’ or having little to do was mentioned in several studies.^{23, 24, 27, 41, 51, 54, 59, 68, 80, 82, 83, 91} Patients suggested that inactivity slowed the in-patient care pathway,⁵⁹ reduced self-efficacy,⁴¹ exacerbated symptoms⁸⁰ and was related to aggression and violence on the ward.²³ Some patients reported that inactivity encouraged poor health outcomes, e.g. saying that they would eat, sleep or smoke but not exercise.^{24, 59, 80, 83}

Authentic experiences of patient-centred care

The final theme brought together a collection of sub-themes focused on authentic experiences of patient-centred care, which included shared decision-making, sensitivity to gender and culture and the provision of information.

Shared decision-making

Two studies reported that patients’ involvement in treatment decisions was associated with positive experiences of care.^{50, 65}

Gender and cultural differences

Patients wanted to be understood and seen as individuals, and this was framed in respect of their gender, ethnicity and religion.^{33, 34, 68, 78} Some patients described cultural differences in perceptions of privacy, and reported concern that staff had not recognised or responded to their discomfort in accepting care from differently gendered staff,⁶⁸ for example during restraint and sedation,³³ or for women with a history of sexual abuse by male perpetrators.⁷⁸ More positively, female patients tended to prefer single-gender wards (where they felt safer³⁶). Where this was not available, female patients were satisfied on mixed wards if they had access to a quiet room, if their privacy was respected and if they had access to personal hygiene products.⁸¹ Faith also mattered: prayer and rituals (e.g. hand washing) offered comfort to some patients⁶⁸ but were not always understood or accommodated by staff.³⁴

Provision of information

There were several reports in which patients felt they had not received sufficient information about their diagnosis,^{23, 65, 69, 87} treatment,²⁰ treatment plan,^{23, 32, 52, 57, 60, 65, 69, 87, 88, 90, 91} choices or rights.^{20, 46, 53, 64, 86} Timing was also important as patients found it difficult to understand or remember this information when unwell.^{45, 69}

Discussion

The aim of this review was to identify the most salient aspects of in-patient experience to support improvements in care in ways that are conducive to recovery-focused care. To the best of our knowledge this is the largest review of its type in the UK and internationally, with 72 included studies, of which a third were from the UK. A strength of the review was the involvement of the PPIRG who provided important face and content validity checks and were able to identify additional areas of experience, such as boredom, which could be built into the main review.

The review makes an important contribution to the field of mental health in-patient experiences through the identification of four key, interlinked themes: the importance of high quality relationships; averting negative experiences of coercion; a healthy, safe and enabling physical environment and ward milieu; and authentic experiences of patient-centred care. These themes and their associated sub-themes represent the active ingredients of a high-quality mental health in-patient experience (as well as the common causes of very poor experiences). The identified themes can be used to design and deliver high-quality services, provide content for the development of robust patient experience questionnaires or inform qualitative methods that aim to evaluate salient aspects of patient experience. They provide evidence for the development of practice guidance that supports the implementation of high-quality services.

A consistent thread across all four themes was the key role of staff in facilitating a high-quality patient experience. However, staff operate within the context of a wider system that needs to support the delivery of care. It was not always possible to understand this context from the studies reviewed as many did not provide such wider contextual information. This would have been useful, particularly in understanding why some studies reported very negative experiences and others reported more positive experiences. Future studies might consider reporting contextual information to aid interpretation.

It is important to note that the findings of studies relating to discharge appeared to be influenced by the research design, with questionnaires identifying high levels of satisfaction whereas experiences captured using qualitative methods were described differently. Future studies should pay careful attention to the way in which design might affect the reporting of experiences.

Limitations

A limitation of this review, common to all secondary research, is that it is reliant on the conduct and content of primary studies which may have included biases that we could not account for. Few studies mentioned the involvement of patients in data collection^{20, 39, 46, 79} and research design,^{20, 27, 39, 46, 79} and the study authors’ professional perspective is often unreported. It is therefore unclear to what extent a study finding reflects the patient voice or whether it predominantly reflects the researchers’ interpretation of their data. Ensuring greater clarity about whose voice is represented, as a means of minimising bias, represents an important methodological challenge for future research. In future reviews, the case could be made to focus on studies where there is evidence of a strong patient voice in the conduct and interpretation of the study.

Although we used the concept of data saturation to decide when to stop data extraction, it is always possible that other papers contained nuances in themes that were unintentionally omitted. The risk of bias in this review may have been partially mitigated by our scoping review which identified key authors and included a citation search of their papers and other literature reviews. In addition, the PPIRG provided important assurance of face and content validity.

Our study relies on secondary analysis of qualitative data. The findings we have presented are drawn from the reports from participants in primary studies. Many of these claims (e.g. the perceived role of good relationships in reducing a range of unwanted outcomes, the role of boredom in exacerbating those outcomes) are reported across multiple primary sources. However, an important limitation of secondary research involves the gaps that exist in studies. A key gap in this review was the lack of experiences from people of Black and minority ethnic groups, which appears to be under-researched. Future studies should ensure they build ethnicity into their design.