Introduction

Mechanical restraint is one of a number of coercive measures used in psychiatric wards in a number of countries [1]. These are commonly applied when there is a need to (i) prevent self-harms or risk of physical injury to the patient, (ii) the staff may be at immediate risk of harm, or (iii) as a measure to prevent dangerous, threatening or destructive behaviour [2]. Typically, restraint is used when a patient becomes violent towards other people or himself. In the case of mechanical restraints, the patients’ arms and legs are physically strapped to a bed by nursing staff (following physicians’ order), restricting use of these limbs. Mechanical restraint has been highlighted as an ethical issue within psychiatric wards [3]. Specifically, the debate revolves around the rights of mentally ill individuals [3] as well as the effectiveness of restraints compared to alternatives which may be experienced as less traumatic on the patients’ part [4]. There is growing international impetus to reduce and eliminate restrictive practices in psychiatric settings [5, 6]. Accordingly, reducing use of coercion measures to a minimum is a highly prioritized matter in health politics worldwide [5, 7].

Currently, reports indicate wide variability in the prevalence of restraints across countries [8, 9]. For example restraints are used in 3.8% to 20% of cases in different settings [10], and describe significant between-country variability in the type and length of restraints during hospitalization [1].

Many possible factors have been suggested to influence differences in practicing restraints across countries, hospitals and wards. Differences in ward culture, treatment ideology, composition of patients, size of wards and number of staff per patient are some of the factors mentioned. Staff attitudes are often mentioned as a possible influence on the use of coercive measures [11,12,13,14,15]. Though individual studies in other non-psychiatric fields (e.g., geriatrics) indicate that participants who experienced or witnessed restraints demonstrated a less judgemental and less negative attitude toward restraint than those who did not [16], this association has not been explored in general psychiatric wards. This may be particularly important as mental health care entails treatment by an interdisciplinary team. Team members may hold very different views regarding restraints which may be affected by the exposure they have had with incidences of restraints.

Accordingly, the aim of the current study was to explore differences in attitudes towards mechanical restraint in psychiatry based on level of exposure of staff members to incidences of restraint.

Methods

Setting

The study was conducted in a government psychiatric hospital in the center of Israel, serving a catchment area of 500,000 residents. The hospital has a 260-bed capacity with 4 acute inpatient units and 4 chronic units, each housing approximately 35 patients.

Participants and Sampling

All members of the medical, para-medical (nurses, psychologists, social workers, occupational therapists), and auxiliary staff were requested to fill anonymous questionnaires over a two-week period (n = 300) from 21/08/2016–07/09/2016.

Questionnaire

All participants completed a sociodemographic and professional questionnaire which included questions regarding gender, age, country of origin, profession, clinical unit and number of incidents in which they were present and participated in incidences of mechanical restraint. In addition, all participants completed a structured questionnaire regarding mechanical restraint adapted from Gelkopf et al. [17]. The questionnaire includes 15 items in five categories: security and care; humiliation and offending; control; order; education and punishment as Table 1 showed. Participants were asked to rate their degree of agreement with each item on a 5-point Likert scale ranging from 1 (do not agree) to 5 (fully agree).

Table 1 Restrains questionnaire (categories and items) and Cronbach Alpha
Full size table

Study Procedure

This study was approved by the Institute’s Research Ethics Board (REB). All participants were provided with information about the study and all questionnaires were anonymous.

Statistical Analyses

For each category, reliability was determined using Cronbach’s alpha. Sociodemographic and professional characteristics between groups were compared using chi-square statistics. Participants were first categorized into two categories based on presence in incidences of mechanical restraints (present or not present) and second into three categories based upon their presence and participation in incidences of mechanical restraint: Not Present (NP), Present Not Participating (PNP) and Present Participating (PP). Comparison of attitudes among the groups was performed using one-way analysis of variance (ANOVA). Tukey’s post hoc analyses were conducted to compare pairs of categories for the three levels of participation. All analyses were conducted using the R statistical package.

Results

A total of 143 people responded to the survey (response rate = 48%). Of these, 46 (32%) reported they were not present during an incident of restraint during the past year (NP) and 97 (68%) were present. Among those present, 35% did not physically participate in restraint (PNP) and 65% were present and participated (PP).

Sociodemographic and Professional Characteristics

The majority of total participants were nurses (57%), followed by paramedical staff (24%), physicians (15%). Furthermore, the majority of participants were women (58%) in the age category of 35–55 years (68%). The vast majority of participants were Jewish (84%) and born in Israel (57%). Among all participants, 34% reported working in the profession less than 10 years, 25% reported working between 10 and 20 years and 41% reported working in their profession for more than 20 years. It should be noted that despite the predominance of women, when comparing characteristics across groups of presence and participation in incidences of restraint the PP group was predominately male (χ2 = 31.125, p < 0.001). Aside from this difference, no significant differences were found in sociodemographic and professional characteristics between the groups.

Attitudes Based on Presence in Incidences of Restraint

Staff members present in incidences of restraint generally reported significantly higher scores in the restraint as security and care category compared to those not present (3.98 ± 0.63 vs 3.70 ± 0.61, respectively, p < 0.05). Compared to those who were not present during restraint, staff members who were present agreed significantly less with statements indicating that restraints are humiliating and offending (3.04 ± 0.55 vs 3.29 ± 0.55, p < 0.05 for those present and not present, respectively). No significant differences were found between the groups in rating categories of control, order and education and punishment (Table 2).

Table 2 Scores on categories measuring respondents’ attitudes towards mechanical restraints based on presence in restraining events
Full size table

Attitudes Based on Presence and Participation in Restraint

Compared to those in the NP and PNP groups, respondents in the PP group reported significantly higher scores on the restraint as security and care category (F(2136) =6.092, p < 0.01) and restraint as order (F(1138) =7.26, p < 0.001) categories. Post Hoc analyses revealed that for the restraint as security and care category the individuals in the PP group reported significantly higher scores compared to the NP group (4.1 ± 0.65 vs 3.7 ± 0.61, respectively, p < 0.05). In the restraint as order category the PP group scored significantly higher compared to the NP group and the PNP group (2.75 ± 0.95 vs 2.17 ± 0.91 and 2.06 ± 0.73, respectively, p < 0.05). Compared to the NP and PNP groups, the PP group reported significantly lower scores in the restraint as humiliation and offending (F(2, 137) =6.231, p < 0.01) category. Post Hoc analyses in this category revealed that individuals in the PP group reported significantly lower scores compared to those in the NP group and the PNP group (2.94 ± 0.50 vs 3.29 ± 0.58 and 3.24 ± 0.58, respectively, p < 0.05). (Table 3).

Table 3 Scores on categories measuring respondents’ attitudes towards mechanical restraints based on level of presence and participation in restraining events
Full size table

Discussion

The purpose of this study was to explore differences in attitudes towards mechanical restraints in psychiatry based on the level of exposure of staff members to incidences of restraint. Our findings indicate significant differences in attitudes towards restraint based on the proximity of staff members to incidents of restraint. Staff members who were present in incidences of restraint saw them more as a means to achieve security and care and less as humiliating and offending. Among those present in incidences of restraint, those who physically participated saw them more as a means to achieve security and order and less as humiliating and offending compared to those present but not physically participating in restraint.

Psychiatric staff members commonly see mechanical restraint as a necessary last-resort intervention to maintain safety of patients [18]. Restraint may be perceived as a manner of protecting patients, and psychiatric staff members may be more likely to use coercive measures in this context [19, 20]. Psychiatric staff members further agree that the use of coercion measures is increasing the safety of staff and others [20]. Gelkopf et al. [17] found that most psychiatric nurses indicated that the main reason for restraint patients is the high risk of self-harm to the patient or injury to the staff and/or environment. In addition, nurses may, on occasion, tend to request restraint patients to keep the ward in order. Results of this study indicate that the level of exposure affected staff members’ attitude towards mechanical restraints. This is in line with studies indicating a correlation between attitudes toward coercive measures and personal experience, reporting that staff members who had been personally involved in implementing coercive measures expressed greater approval of the use of them [14, 21]. Similarly, Gelkopf et al. [17] noted that reasons for restraint a patient were affected by whether nurses worked in either closed or open wards. Though not directly explored in this study, given the nature of patients in each type of ward, it is probable that nurses who work in closed wards were more exposed to incidences of restraint, affecting attitudes.

Our findings may be partially explained by the feelings of fear and personal concern associated with physically participating in restraint. Psychiatric staff members are at increased risk of physical injury, emotional effects, and death [22]. Foster et al. [23] reported that medical staff members were more likely to use physical methods to manage incidents of aggression when they experienced fear, which can be induced by working in environments such as psychiatric wards. Wards with higher rates of aggression and insufficient safety in the workplace have been found to have a preponderance of nurses whose style of interacting and intervening is restrictive and controlling [19, 24]. Psychiatric nurses reported being faced with a decisional dilemma in situations which potentially required consideration of physical restraints. Their decision was complicated by the potential for harm, unsuccessful search for alternatives, choosing among equally unwelcome options, and unit policies and professional attitudes [25]. These may all contribute to the understanding of our findings.

Staff members who were present and physically participated in restraint were more liable to report that restraint is a way to achieve order in the ward (e.g., stopping a brawl, separating fights, etc.) compared to staff members not present or not participating in restraint. This is in line with previous reports indicating that nurses working in closed wards (naturally more exposed to incidences of restraint) considered restraint as a means to separate fighting patients and stop a brawl more than those in open wards [17].

Our findings indicate that staff members who were present and physically participated during restraint are less likely to believe that restraint is degrading and emotionally harmful to patients compared to staff members who were not present or did not physically participate in restraint. Mental health nurses who care for and support distressed and/or disturbed inpatients in acute psychiatric settings may experience substantial emotional quandaries and concurrent cognitive dissonance [26]. The cognitive dissonance is discomfort experienced when simultaneously holding two or more conflicting cognitions, ideas, beliefs, values or emotional reactions [27] when mechanically restraint patients. Cognitive dissonance is the distressing mental state that people feel when they find themselves doing things that don’t fit with what they know, or having opinions that do not fit with other opinions they hold [18]. This theme highlights ethical concerns and internal conflicts expressed by nurses attempting to balance rights to liberty and dignity and the protection of the self and others [28]. Staff members in psychiatric wards may feel socially pressured to “control” patients who have lost their rational capacities [29] and those who consider seclusion and restraint demeaning to the patient and contrary to the principles of autonomy and care may feel conflicted [19]. Previous studies showed that a considerable proportion of staff members were not very critical of the use of coercion and did not think of coercion as offensive towards patients. Perhaps the idea of providing good care excludes the thought that its use can also be offensive and potentially harmful towards patients [17, 30]. In contrast to these attitudes on the part of some mental health practitioners, Kinner et al. [20] reported that patients were more likely to perceive restraints as harmful. This discrepancy in attitudes may add to existing tension in psychiatric wards.

A Cochrane review conducted by Sailas & Fenton [31] concluded that no clear recommendation can be made about the effectiveness, benefit or harmfulness of seclusion or restraint given the absence of any controlled trials in those with serious mental illness. Nevertheless, qualitative studies have reported substantial deleterious physical and, more often, psychological effects of restraint on both patients and staff [32]. Restraint patients in whichever way may cause both emotional and cognitive problems for staff [28]. This requires careful consideration of restraint practices, and several countries have invested resources into establishing alternative manners of achieving safety.

Limitations of this study should be taken into consideration. Presence and participation in restraint was based on self-report, which lend itself to report and recall bias. Nevertheless, given the sensitivity of the issue at hand, we found it important to include anonymous questionnaires in order to increase truthful responses. Second, the response rate is below the desirable 70%, which may affect generalization of our results.

Conclusions

This study highlights an important aspect commonly overlooked in previous studies regarding restraints in psychiatry. Attitudes of staff members towards restraint differed based on the level of presence and participation in the physical act of restraint. The greater the proximity of staff members, the more they reported perceiving restraint as a manner of achieving “security and care” and less “humiliating and offending”. This may have implications in understanding the professional and social discourse concerning mechanical restraints. In order to reduce the use of mechanical restraint, the conflicting feelings of staff members about use of restraint should be addressed and alternatives such as de-escalation approaches should be applied in organizations. Research regarding additional factors affecting attitudes towards mechanical restraint in psychiatry is needed.