Attitude and practice toward physical restraint among psychiatric nurses in Guangdong, China: a cross-sectional survey^†

This study was conducted via an online anonymous survey performed from November 1, 2018 to December 31, 2018.

The convenient sampling method was used to recruit psychiatric nurses in Guangdong, China. The inclusion criteria were (1) registered psychiatric nurses or licensed practical psychiatric nurses and (2) Chinese speakers. Nursing students were excluded.

Data were collected through a self-designed questionnaire, which includes participants’ personal characteristics and Physical Restraint scale (PRS). The questionnaire was adjusted by researchers to make it easier for participants to understand. This self-designed questionnaire was tested by four nursing experts (faculty members from the nursing department and clinical nurses) who have >20 years of working experience in the psychiatric nursing setting.

After seeking approval from the nursing director of each psychiatric hospital that is registered in Guangdong Nursing Association (a non-governmental organization), the link of online-questionnaire was sent to the mailbox of the nursing administration department. The survey purpose, significance, and instructions were provided to ensure a high participation rate and obtain valid data. The nursing director of each psychiatric hospital helped the researchers to spread the questionnaires using WeChat app (a popular social application in China). All potential participants accessed the questionnaires and answered individually using WeChat via smartphones. Informed consent was presented on the first page of the online survey and was obtained before participants completed the online survey. To avoid repetition, a WeChat account was allowed to submit the questionnaire only once.

Data were collected online through the Chinese survey website named Wenjuanxing (www.sojump.com) from November 1, 2018 to December 31, 2018. A total of 1009 questionnaires were collected. The collected data were double-checked by two researchers (Cai and Xiao). Ultimately, 356 questionnaires were excluded because: (a) approximately 100 respondents indicated they worked outside Guangdong; (b) 128 questionnaires were with contradictions (for instance, the number of working years being in excess of age, age failing to match the professional position); and (c) 128 questionnaires were filled with identifiable errors. The number of the included qualified questionnaires were 753, accounting for 74.6% of all returned questionnaires.

SPSS 21.0 Software Package was employed to analyze the data. Attitude and practice on PR use showed abnormal distributions (as examined by the Kolmogorov–Smirnov test). Descriptive statistics (frequency and percentage) were used to describe the participants’ personal characteristics and profession information. The Mann–Whitney U test and the Kruskal–Wallis test were used to determine whether the personal and professional characteristics had any effect on the scores of attitude and practices, respectively. In addition, ordinal regression analysis was administered to identify the correlation between attitude and practice related to PR. Confidence interval (CI) was set as 95%. The significance level was set to 0.05 (two-tailed).

A pilot study is usually employed to test the design of the large full-scale study, which then can be adjusted.²¹ In this study, the aim of the pilot study was to measure the simplicity of questionnaire and estimate the time needed for taking it. According to Thabane et al.,²¹ the required sample for a pilot study would be at least 75 participants. Therefore, a pilot study was performed with 75 nurses to test the feasibility of the scales and instrument development. Simple modifications of the tools were done and all of these 75 questionnaires were excluded from the data analysis. Sample size was calculated by an online tool available at https://www.calculator.net. The minimum sample size was 385, and we used a larger sample size.

The analysis of 753 psychiatric nurses’ personal and professional characteristics demonstrated that almost half of them were aged between 25 and 35 years (47.9%) (Table 1). A total of 77.0% of the nurses were female and more than half of them were married (63.6%). Approximately half of the nurses (51.9%) had obtained a bachelor’s degree. Most of the nurses had a period of working experience which amounted to <5 years (38.4%), and 75.4% of them were working in closed psychiatric wards. Further, most of them (69.6%) were junior nurses. Only 22.7% of them were frequently given the opportunity to participate in training programs.

Regarding the attitude of psychiatric nurses toward PR use, 17 (2.3%) of them showed scores that are <9, 480 (63.7%) had scores ranging from 10 to 19, and 256 (34.0%) had scores ranging from 20 to 28. The total scores ranged from 4 to 27 (median = 18.00, SD = 3.31), implying a neutral attitude (Table 2). The scores of each item are shown in Table 2. In total, most nurses agreed with the statement that “in emergencies, nurses are allowed to apply physical restraint for patients without a psychiatrist’s order.” About 30.4% disagreed with the statement that “in all cases, the reasons for physical restraint should be explained to the patient/family members and require their informed consent.” About 55.4% of them agreed that “there should be standard process when implementing physical restraint.” Approximately 55.6% agreed with the statement that “physical restraint would cause physical and psychological harm to patients,” and 53.1% agreed that PR could easily lead to nurses’ injury and psychological stress. Only 13% strongly agreed that they had received enough training in terms of PR.

Concerning nurses’ practice in relation to PR, 312 (41.4%) of them showed scores that are <12, 242 (32.1%) had scores ranging from 13 to 16, and 199 (26.4%) had scores ranging from 17 to 24. The scores were in the range of 3–24 (median = 14, SD = 4.49), reflecting a moderate level of practice (Table 2). Totally, 74% of nurses agreed with the statement that “physical restraints would be applied when the patient is at risk of violence/suicide/escape”, and 73.8% reported that they would consider whether alternative methods are adequate before implementing PR for the patient.

Table 3 shows the factors that affect nurses’ attitude and practice regarding PR use. The Mann–Whitney U test demonstrated that marital status, gender, and participating in nightshifts influenced their practice on PR use (P < 0.05). The Kruskal–Wallis test indicated that age, education status, working experience, professional position, and the frequency of training program influenced nurses’ practice of restraint (P < 0.05). The attitude score was influenced by gender (P = 0.001), age (P = 0.002), professional position (P = 0.009), and the frequency of training program (P = 0.000).

Table 4 presents the association between attitude and practice of nurses related to PR. The practice score was set as a dependent variable; the attitude score was set as an independent variable; gender, age, working experience, and professional position were set as covariates of practice. Ordinal regression showed that nurses with a more negative attitude toward PR were more likely to use it (OR = 1.91, P < 0.001).

In this study, respondents demonstrated neutral attitude toward PR. However, some of them still held negative attitude toward PR. For example, about 33% disagreed with the statement that “in all cases, the reasons for physical restraint should be explained to the patient/family members and require their informed consent.” This result of the current study was similar to the findings of the Mahmoud study,¹ which revealed that about 34% of participants did not agree that family members had the right to refuse the use of PR. This may be due to the rare involvement of ethical issues regarding PR use in training or nursing program education.²² This study indicates the need to increase the awareness of nurses towards patients’ rights and the ethical issues related to the use of PR to protect patients’ autonomy and rights.

In addition, in our study, about 76.7% agreed with the statement that “in emergencies, nurses are allowed to apply physical restraint for patients without a psychiatrist’s order.” We found in agreement with other studies that most nurses reported that using PR does not need a doctor’s order.^23,24 Our study revealed that PR was commonly employed and removed according to the nurses’ subjective clinical judgment. Indeed, the decision to apply PR is not an easy one, and the nursing staff revealed that it is an ethical dilemma.⁴ However, it must be noted that if PR is decided to be implemented for patients who lack capacity, their fundamental rights and freedom must be respected and protected to the maximum extent and their best interests should be maintained.¹

Furthermore, only about 40% agreed that they had received enough training in terms of PR, and the level of degree of this item is relatively low. Our study indicates that the training needs of nurses were strong regarding PR. Therefore, it is important to note that nursing managers and leaders of nursing institutions should pay more attention in adopting effective training strategies to effect the nursing staff’s attitude and enhance their understanding of PR. The purpose of this exercise is to inspire nurses to proactively seek alternatives to PR and attempt to minimize the use of PR.²²

Regarding the scores for nursing practice, the results of this study indicate that psychiatric nurses have a moderate level of practice related to PR use. The results of our study revealed that 74% of nurses agreed that PR should be applied when the patient is at risk of violence, suicide, or escape. This finding reflected their agreement on using PR for psychiatric patients. Such a finding could be explained in terms of the fact that that nurses perceived agitated psychiatric patients as being dangerous, and they agreed to use PR to reduce the patients’ dangerous behaviors to ensure safety. In addition, it may be explained that nurses tend to be more likely to use PR for the convenience of their own work, especially when the nursing assignment is heavy and supervision capacity is limited.²⁵ Further study supporting this evidence was found by Mahmoud¹ and Al and Hasan.¹⁰

However, concerning finding alternative approaches, 73.8% of the respondents reported that they would consider whether alternative methods are adequate before implementing the PR for patients. In contrast to some other studies,^13,24 the majority of our participants had shown awareness concerning alternatives to PR. The results of our study are consistent with the Mahmoud¹ findings, which showed that 94.7% of the participants would try alternative nursing measures before restraining patients.¹ Through standardized training, nurses would be able to apply PR more rationally. Nevertheless, in the current study, there were still about 30% of participants who demonstrated that they failed to find alternatives to PR. Therefore, introducing and focusing on alternatives to PR would be a vital topic in restraint minimization training programs.

Finally, in our study, approximately 21.5% of the respondents agreed that PR could be applied for the convenience of nursing work, and junior nurses could be recommended to use PR appropriately for the convenience of work. The findings of this study revealed that part of the nursing staff misused PR. This, to some extent, may be due to the workforce shortage in psychiatric institutions in China. An inadequate number of psychiatric nurses care for a considerable number of psychiatric patients; this leads to their heavy workload, which is considered to be one of the critical reasons of the prevalent use of PR.⁴ However, through these findings we hope to offer evidence for nursing managers and policy-makers who are developing targeted intervention strategies of regulating the use of PR that would benefit clinical practice.

This study found that nurses within the age group of 35–45 years had much higher attitude scores than the nurses from other age groups (χ² = 14.995, P = 0.002). This finding indicated that psychiatric nurses were influenced by the experiences that they have learned during the period of their employment as a psychiatric nurse. Those nurses who were aged between 35 and 45 years were the most senior nurses, possibly because senior nurses were more experienced, and continuous training to some extends help to change the nurses’ attitude toward PR. This finding is generally in line with Wang et al.²² and Mahmoud.¹

The present study revealed that nurses aged >45 performed practice inadequately, whereas nurses aged below <25 years performed better, from which it was found that the level of nurses’ practice concerning PR decreased with older nurses (χ² = 57.849, P = 0.001). These findings are inconsistent with the study by Mahmoud¹ which concluded that experienced professionals’ age-wise, older nurses tended to make better adjustments than younger nurses. This is mainly related to the decrease of frequency of training in PR among these older-age nurses. Furthermore, the male nurses have higher attitude and lower practice scores (Z = −3.436, P = 0.001; Z = −3.622, P = 0.001), and this reveals that male nurses showed better attitude and performed better skills in relation to PR.

It is interesting to note that nurses who received more than 6 times the length of training program related to PR showed better attitude and performed better than those who received <4 times the length of training program (χ² = 35.002, P = 0.001; χ² = 11.227, P = 0.004). This finding, to some extent, reveals that implementing a training program related to PR would be an effective approach in enhancing nurses’ attitude and improving their practice in relation to PR. There were no significant differences between nurses’ other demographic and professional characteristics and attitude and practice scores in using PR. In brief, each psychiatric institution can find out the weak links in the institution according to the above affecting factors that influence the use of PR and improve the awareness of PR of psychiatric nurses through different training programs.

In harmony with the conclusions reached in the literature of Suen et al.¹³ and Werner,¹⁶ this study found that the practice related to PR was associated with nurses’ attitude toward PR use. Specifically, through ordinal regression analysis, we found nurses with a more negative attitude toward PR were more likely to use it (P < 0.001). “Negative attitude” means the nursing staff in the study of Suen et al.,¹³ the path analysis indicated that nursing staff’ attitude had positive direct effects on PR use, which was similar to the result of the current study. Likewise, in their study, Werner¹⁶ found after using regression analyses that participants’ attitudes were significantly associated with intention to use PR. In addition, our findings support the conclusions of other researchers that the nursing staff’s attitude and beliefs are important determinants of PR use.^10,13 In the study by Eskandari et al.,²⁴ multiple linear regression showed that only 6% of variations in nurses’ practice was explained by participants’ attitude and intention concerning PR. The reason for their result could be that, although nurses’ attitude and intention were influencing factors of nurses’ practice, other factors could have existed to affect nurses’ practice regarding PR. However, in our study, to identify the correlation between nurses’ attitude and practice, we employed the ordinal regression model to control the factors influencing practice regarding PR use, to a greater extent showing that nurses’ attitude can affect the practice of PR.

Therefore, we proposed that nurses’ attitude was the main determinant of intention to use PR in psychiatric settings. There are two major possible explanations regarding this finding. First, from the theoretical perspective, similar to the findings of the other study using the TRA with nurses,¹⁶ it may be proposed that nurses’ personal beliefs and attitude about the use of PR are more potent in determining their intention when compared with other factors. The second explanation may be related to the environmental and cultural background of this study. Although there currently exist guidelines governing the application of PR in relation to mentally ill patients in China, there is still a lack of coercion supervision from the nursing administration department or the third party with regard to PR use in psychiatric settings.⁴ Thus, the intention to use PR may be mainly influenced by nurses’ personal attitude rather than organizational factors or ward culture.