Abstract
Objective:
This systematic review presents evidence regarding factors that may influence the patient’s subjective experience of an episode of mechanical restraint, seclusion, or forced administration of medication.
Method:
Two authors searched CINAHL, PubMed, SCOPUS, Web of Science, and Psych-Info, considering published studies between 1 January 1992 and 1 February 2016. Based on the inclusion criteria and methodological quality, 34 studies were selected, reporting a total sample of 1,869 participants.
Results:
The results showed that the provision of information, contact and interaction with staff, and adequate communication with professionals are factors that influence the subjective experience of these measures. Humane treatment, respect, and staff support are also associated with a better experience, and debriefing is an important procedure/technique to reduce the emotional impact of these measures. Likewise, the quality of the working and physical environment and some individual and treatment variables were related to the experience of these measures. There are different results in relation to the most frequently associated experiences and, despite some data that indicate positive experiences, the evidence shows such experiences to be predominantly negative and frequently with adverse consequences. It seems that patients find forced medication and seclusion to be more tolerable than mechanical restraint and combined measures.
Conclusions:
It appears that the role of the staff and the environmental conditions, which are potentially modifiable, affect the subjective experience of these measures. There was considerable heterogeneity among studies in terms of coercive measures experienced by participants and study designs.
The current models and paradigms in mental health care are oriented towards the goals of recovery and self-determination for the user, and explicitly propose the reduction of coercive measures in the treatment of severe mental disorders. Forced medication, seclusion, and mechanical restraint in theory are intended to protect patients and those around them but the use of these measures restricts freedom and conflicts with the ethical principle of patient autonomy. These coercive measures also present risks due to their side effects’; for example, forced medication may be associated with hypotension 1 and bronchospasms. 2 These measures are even more problematic when used repeatedly 3 and in combination—for example, when seclusion or forced medication is combined with physical or mechanical restraint, which can become extremely traumatic, causing physical and psychological harm to both patients and staff, 4 and leading to serious side effects, including death. 5
The use of coercive measures is controversial due to the lack of consensus regarding the choice of coercive method, its practice, registration, and control, and the core issue remains as to which is the best intervention(s) to prevent the application of these measures. 6 –8 Reduction rates in restraint have been achieved through multimodal interventions, 9 which include several strategies to prevent and control disturbed behaviors without using restraint measures. Decreases in seclusion-restraint rates have been found in different institutions and countries that have applied the 6 core strategies: 10 –14 improving leadership, training staff, monitoring the use of restraint, involving consumers and family, using tools to reduce episodes of seclusion or restraint, and analyzing past events. In addition, the intervention, based on the “safewards” 15 model developed by Bowers, has shown a significant reduction in restraint measures in a recent cluster randomized trial. 16 Nevertheless, there are few studies on patient experiences of these measures and the factors that influence there experiences.
In recent years, the shared decision making process and the involvement of patients in their treatment have emerged with great force in the healthcare system, with reference to the process of taking control and responsibility for life, incorporating the goals of self-sufficiency, dignity, respect, and membership in and contribution to the community. 17 It is at this point that the subjective experiences of the patients about their treatment after experiencing a coercive measure acquires a special importance, since it represents a central theme regarding the quality of the attention that mental health services are trying to offer. 18 Some qualitative and quantitative studies have explored the subjective experiences, and related variables, of the use of coercive measures on patients admitted to mental health hospitalization units. 19 –22 Although there are other reviews on the subjective experiences of patients after the application of coercive measures, 23 to our knowledge, no integrated results have focused on the factors that influence this experience. An understanding of the factors affecting the subjective experience of patients may help to improve the functioning of psychiatric units and thus improve patient satisfaction with the treatment. This understanding could also eventually contribute to improving programs aimed at preventing the use of coercive measures.
The objective of this study was to carry out a systematic review of the factors that influence the patient’s subjective experience in relation to the application of coercive measures in mental health hospitalization settings. The secondary objective was to review the results concerning the types of subjective experiences reported by the patients and the outcomes, comparing the different coercive measures and considering the patient’s subjective experience.
Methods
This review followed the PRISMA guidelines 24 and the protocol was registered in PROSPERO (registration number: 42016036669).
Eligibility Criteria
Participants
The review considered only: 1) studies carried out in samples of patients exposed to physical or mechanical restraint, seclusion, or forced medication; 2) studies where the coercive measure was applied during psychiatric hospitalization; and 3) studies with a qualitative or quantitative assessment about the subjective experience in relation to the coercive measures selected. Studies were excluded when: 1) patients were under 18 y old; 2) patients were admitted into non-mental health wards, specialized facilities for substance abuse, or psychogeriatric facilities; or 3) the study only accounted for the experience of the staff.
Intervention or exposure
We considered for inclusion studies that collected information of the factors associated with the subjective experience of patients exposed to mechanical restraint, seclusion, or the forced administration of medication.
Outcomes
The primary outcomes were factors that affect the patient’s subjective experience of the coercive measures, considering subjective experience to include all those perceptions, feelings, perspectives, impressions, or opinions that the patients have with respect to the coercive measures. The secondary outcomes were: 1) the types of subjective experiences reported by the patients, and 2) the different measures (mechanical restraint, seclusion, and forced medication) considered comparatively in relation to the patient’s subjective experience.
Search Strategy and Selection of Studies
The aim of this systematic review was to identify studies published between 1 January 1992 and 1 February 2016 in the English or Spanish languages. A 25-y time period was chosen, as it was considered to be sufficiently broad and representative for the purpose of the review. We searched the following databases for relevant articles: CINAHL, PubMed, SCOPUS, Web of Science, and Psych-Info. We used the following subject headings and search terms to identify potential articles for inclusion: “mental health or psychiat* (in all fields) AND restraint or seclusion or isolation or involuntary medication or forced medication or coercion or coercive measures (in all fields) AND patient experience or patients experiences or patient perspective or patients perspectives or patient perception or patients perceptions or patient preference or patients preferences or satisfaction (in all fields).”
Study selection was carried out independently by 2 reviewers (CAS and JGP) between March and September 2016. A second review was conducted between August and September 2017. The results of the search were first screened by title and abstract, and where doubt existed, the complete text was revised. Studies that met the inclusion criteria were then selected. If there were discrepancies for the inclusion of any study, an agreed decision was made between the 2 reviewers. After our initial search, we also manually reviewed the reference lists of selected articles. The details of the search, including the reasons for exclusions, are presented in Figure 1.

Flow chart of included studies in the systematic review.
Methodological Quality and Data Extraction
To ensure the quality of the studies included in the analysis, we used Critical Appraisal Skills Programme (CASP), 25 a standardised checklist on the most relevant characteristics of different types of studies that provides a score for each study. For qualitative studies, the scale presents 10 items; for case-control studies, 11 items; for longitudinal studies, 12 items; and for randomized control trials, 11 items. We gave a “high quality” rating to studies with ≥70% of items satisfied; “medium quality” to studies with ≥50% and <70% of items satisfied, and “low quality” to studies with <50% of items satisfied. Any discrepancy between the reviewers was resolved through discussion. All the selected studies passed the quality assessment.
Data were extracted independently from each of the 34 studies by 2 reviewers (CAS and JGP) regarding: 1) country of origin, 2) setting, 3) sample, 4) coercive measure used, 5) design and methodology, 6) the factors that affect the experience of the measure, 7) the subjective experience associated with the measure, and 8) comparison between measures.
Data Synthesis
Given the diversity of studies in terms of design, type of sample, type of coercive measure used, and scope of study, the narrative method was considered the most appropriate procedure by which to present the results. The data were grouped by thematic categories. Quantitative data were also included in the data synthesis.
Results
Characteristics of Included Studies
Thirty-four articles that met the inclusion criteria were identified and passed our quality assessment as high-quality studies (see flow chart in Figure 1). The studies included a total of 1,869 participants. Most of the studies were set in different acute psychiatric wards in hospitals (28), as well as in forensics wards (2), community mental health services (regarding the experience during hospital admission) (2), a mixed setting (inpatient units in a general hospital and forensic wards) (1), and a long-term care organization (1). The distribution by country was as follows: United States (8), Canada (4), Netherlands (4), Australia (3), Germany (3), Sweden (2), United Kingdom (2), Finland (2), Norway (2), South Africa (2), China (1), and Austria (1).
In relation to the typology of the study, 15 used qualitative methods, with structured or semi-structured interviews (14) and the auditing (1) for data collection. Another 13 studies were quantitative, of which 8 were cross-sectional, 4 longitudinal, and 1 an intervention study. The methodology of the remaining 6 studies was mixed (qualitative and quantitative). Regarding the types of coercive measures used by the staff, 16 studies used seclusion, 5 studies used mechanical restraint, and 1 used forced medication as the only measure. Eight studies took into account 2 measures, and 4 studies included the use of 3 measures. More information is summarized in Table 1.
Details of Included Studies
Factors Affecting the Subjective Experience of the Coercive Measures
The data were reduced and categorised according to the following categories: 1) Provision of information, presence of or interaction with staff, and adequacy of communication with professionals; 2) the physical environment of the psychiatric ward; 3) respect, humane treatment, and support from staff; 4) debriefing; 5) Individual characteristics of patients and their admission or treatment.
Provision of information, presence of or interaction with staff, and adequacy of communication with professionals
Twenty studies indicated that the provision of information, the presence of or interaction with staff, and the adequacy of communication with professionals influenced the subjective experience of the coercive measure(s). Ten of these studies indicated that contact and the presence of staff during the process influences the subjective experience, making the coersive measure less aversive. 26,30,33,37,42,43,45,48,49,54 One study 49 indicated that the presence of relatives was associated with a more positive experience, and 2 other studies mentioned that staff trying to calm or solve the user’s problems before the application of the coercive measure also had a positive influence. 35,50 Seven studies indicated that providing adequate information about the measure 26,34,42,48 and making sure that the patient understands the reasons for adopting the measure, 38 or the staff’s failure to explain them, 36,39,47 affected the experience. Furthermore, adequate communication 41,44,49 or a lack of communication or communication skills 27,28,39,47 was also noted as a factor that may influence the experience of coercive measures in 7 studies.
The physical environment of the psychiatric ward
Eleven studies cited the physical environment as an influential factor in the subjective experience of coercive measures. This concept encompasses a range of concepts, as indicated by Van der Schaffer et al., 56 including the comfort of the rooms 29,33,34,37,54,55 and/or furniture, 33,37 the physical environment, 29,33,34,55 wearing own clothes, 37,45,54 the presence of personal objects, 29,30 the regulated use of the bathrooms, 33,37 pronounced inactivity, 33,48 the ward atmosphere, 47 noise, 48 hostile environment, 34 and privacy. 48 In one study, 40 it was found that even the physical environment not directly related to the measure may influence the perception of seclusion. The authors further suggested that seclusion was associated with a positive experience in patients who stayed in rooms with multiple beds, and the authors speculate that the previous discomfort and lack of privacy could be the reason why seclusion has been frequently experienced as positive.
Respect, humane treatment, and support from staff
Ten studies reported that respect, humane treatment, and support from staff influenced the experience of the measure, with several terms related to a better experience of the measure: respect, 35,49 professional and pleasant behavior of the staff, 26,39 humane treatment, 33 treatment fairness, 21 taking into account the dignity of the patient, 34 empathy, 41 a position at an equal level, 46 and support from the staff. 35
Debriefing
In 8 studies, 20,22,26,36,42,45,48,55 the use of debriefing was indicated as an important technique for reducing the emotional impact of the measure. In 4 of these studies, 26,48,45,55 patients reported their desire to talk about what happened as a way of processing the experience. In 3 studies, 26,48,36 debriefing was suggested to prevent subsequent trauma and distress. However, one of the methodologically more comprehensive studies revealed that debriefing after seclusion did not significantly reduce the trauma experience. 31 The study by Ling et al. 20 indicated that debriefing facilitated a greater understanding of restraint events, with the same conclusion reached by Holmes el al; 22 albeit, their data was from the staff reports. Finally, the study by Wynn et al. 42 recommends debriefing to improve patients’ lack of understanding of the reasons why the measure was adopted.
Individual characteristics of patients and their admission or treatment
A total of 7 studies mentioned that characteristics at the level of the individual, including their admission and treatment, were associated with a negative experiences of the coersive measure. In one study, 51 patients with low insight into the disease perceived coercion more strongly than patients with high levels of insight. In another study, poor adherence to treatment and multiple previous episodes of coercive measures 53 were also correlated with a negative experience. One study 32 found that a younger age and female gender were associated with more burdensome coercive measures. Other factors were associated with more positive experiences, including psychotic symptoms, 42 voluntary admission, 35 and previous admission in the unit. 52 Yet, a history of substance abuse was related to a negative experience, 53 and the consumption of prescribed medication had both positive and negative effects. 19 Finally, one study 32 found that application of combined measures was associated with a higher perception of coercion.
Characteristics of Experiences Raised by Coercive Measures
Twenty-six studies reported predominantly negative experiences associated with the measures (irrespective of the measure applied). 19 –22,26 –28,30,31,33,34,36,37,39,42 –50,52,54,55 The most frequent experiences reported were: 1) fear, anxiety or post-traumatic stress disorder (PTSD) 19,20,28,30,31,33,34,41,42,46 –49,51,53,55 ; 2) powerlessness, abandonment, distrustfulness or loneliness 20,22,26,28,30,33,34,38,42,45,47 –49,55 ; 3) punishment, maltreatment, or pain 27,28,36,37,39,44,46,48 –50,53 ; 4) anger, rage, or resentment 20,28,30,33,34,42 –44,46,55 ; 5) depression, impotence, or sadness 19,30,43,44,46,51,54,55 ; 6) humiliation, degradation, or shame 34,47,28,29,55 ; and 7) loss of freedom, or coercion. 19,21,29 However, 6 studies found more positive than negative experiences, 35,38,40,41,51,53 and one study showed an equal number of positive and negative experiences 29 associated with the application of the coercive measures. In 14 studies, a minority of patients had some positive experiences after the application of the measures, mainly regarding seclusion. 22,26,27,30,34,37,39,43,44,46,48 –50,54 The most frequent positive experiences reported were: 1) feeling that the measure was helpful, beneficial, or necessary 29,35 –37,44,49 –51 ; 2) calming down, time for reflection or rest 26,27,30,34,38,39,42,44 ; 3) safety or sense of control 26,34,35,38,48,49,54 ; and 4) prevention of violence or a place to express emotion 34,44
Subjective Experiences Comparing the Different Coercive Measures
Eight studies mentioned a comparison of the subjective experiences of the different coercive measures. 19,21,30,32,34,36,38,54 Most of the studies identified forced medication as the preferred coercive measure for patients as compared with restraint, 19,21 or seclusion, 34 with patients finding it less traumatizing and distressing. 32,36 The study by Naber et al. 51 found a tendency toward patients viewing mechanical restraint as more unpleasant than injections but the result did not reach significance. A study, with a high-quality methodology, found in their follow-up that mechanical restraint was associated with a higher perception of coercion and other negative experiences as compared with seclusion. 30 Another study with a high-quality methodology, however, indicated no significant difference in perceived coercion between seclusion and forced medication. 38
Discussion
General Findings
This systematic review provides a detailed exploration of the factors that influence patient’s subjective experiences of restraint, seclusion, and/or the administration of forced medication in inpatient mental health services.
Among the main findings on the factors influencing the perception of measures are the attitudes of professionals and patient’s interactions with the staff. Thus, there is evidence to support the importance of staff behavior, especially their presence, during episodes of hostility and/or agitation and during the implementation of coercive measures, 26,33,37,42,43,48,49,54 positioning them as agents of protection who are in control of the situation. 36,41,42,48,49 Importantly, this result is related to modifiable factors, and reveals the need to have experts who are specialized in mental health care, with adequate training and specific organizational control to ensure good clinical practice in the wards. Indeed, staff training is covered in most programs to reduce coercive measures 8 and it is part of the 6 core strategies. 57 Likewise, respect and humane treatment by staff were repeatedly associated with a more favorable perception of coercive measures, a finding mainly reflected in qualitative studies, 33,37,55,58 with a high empathy rating of staff related to a decrease in the use of coercive measures. 59 Improved relationships between patients and staff seem to reduce conflict and restraint rates, as shown by the “safewards” model. 16 Moreover, positive treatment by staff upon hospital admission was also associated with less coercive experience. 60,61 The use of debriefing after the coercive measure appears to be supported by empirical evidence; 20,22,26,36,42,48,55 although, its effectiveness in preventing trauma is inconsistently reported. 31 However, debriefing plays an important role in providing psychological and operational feedback for both users and professionals, as well as for the organization as a whole. Another important issue that must be considered is the need to create a comfortable and appropriate climate in the ward, also taking into account individual patient characteristics and the possible negative consequences that these measures may on certain patients. Likewise, several studies have shown that the use of sensory modulation 62 or comfort rooms can reduce the need for coercive measures. 56,63 Thus, the evidence seems to indicate that creating a comfortable and appropriate atmosphere in the ward not only improves the subjective experience, but also prevents the adoption of these measures. With regard to the individual characteristics, there are fewer studies and more heterogeneous results. However, the results suggest that patients who are more vulnerable, with more adverse experiences and repeated coercive measures, present a greater risk of possible complications.
Some studies have demonstrated perceived benefit after seclusion 43,44 and/or mechanical restraint, 26,29 with positive perceptions including feelings of protection and security. Nevertheless, most studies reflect adverse and negative experiences, and these measures are associated with post-traumatic stress disorder 19 and a high degree and wide variety of negative feelings and experiences, such as fear and humiliation.
Comparing among coercive measures, the results suggested a less aversive subjective experience associated with the administration of forced medication 19,21,34 and more negative outcomes related to mechanical restraint 30 and combined measures. 32 Nevertheless, while some measures are more acceptable to patients than others, it is important to consider the preferences of patients and, if possible, discuss them on admission.
Limitations
There are a number of important limitations in this systematic review. The main limitation is related to the heterogeneity of the studies, with different designs, the use of both qualitative and quantitative (or both) methods, and, for most studies, small sample sizes. Also, the studies analyzed different coercive measures or combinations of measures. In general, studies about the experiences of coercive measures, as well as studies about the factors that affect them, are methodologically complex, 64 and should be considered in the context in which they were carried out. Variations in results can be attributed to different service settings, mental health legislative frameworks, social policies, and cultural factors, and this requires that the results be interpreted with caution. Second, given the object of such studies and their important ethical and legal connotations, a publication bias is possible; i.e., there could be instances where negative results were obtained and not published. Third, study selection bias is possible; although, 2 independent reviewers performed the selection of the sources used in this study. Finally, the factors that affected the subjective experience were not considered as primary outcomes in most of the studies.
Future Research
Few studies used valid, simple, and reliable assessment instruments to study the subjective experience of coercive measures and the factors that influence these experiences. As pointed out by others, 65,66 there are currently few instruments that can measure the subjective experiences of patients and may be applicable to compare the different coercive measures and different interventions. In this review, there were only 3 longitudinal studies 19,30 and one interventional study 31 with standardized outcome measures. Thus, more longitudinal and interventional studies are necessary to understand what factors positively and negatively affect patient experiences. Although many intervention studies have been carried out to reduce the use of these measures’ and to improve the handling of conflict and incidents in wards, these studies do not address the experiences of patients and the possible improvements in such experiences. Thus, subjective experience tracking and an evaluation of the long-term psychological effects are still lacking in this domain. Furthermore, there are few studies concerning the individual factors associated with the subjective experience of coercive measures, such as those related to previous exposure to traumatic or violent events, which may evoke stressful emotions and feelings among patients. These factors could be identified to prevent possible adverse effects of these measures.
Conclusions
It can be inferred from the results of this review that the actions of staff, the respect and treatment afforded to the patient, the effects of the physical and the organizational characteristics of the unit, and the inclusion of debriefing all influence the experience of coercive measures, and therefore efforts should be made to implement these changes in wards. In addition to reducing coercive measures, it could also be possible to improve the subjective experience of users in cases when the adoption of the measure is inevitable.
Footnotes
Author Note
Joint first co-authorship: Carlos Aguilera-Serrano and Jose Guzman-Parra
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
