One of the most important responsibilities of nurses in care is to prevent harm to the patient and others, and many nurses use physical restraint to achieve this goal [4]. Physical restraint is any device and material that is attached or adjacent to patient’s body such that s/he cannot control or easily remove it. Ankle or wrist restraints, chairs with table trays, or full use of bedside rails are some of physical restraints [5]. Considering that the use of physical restraints in older patients is associated with physical and psychological complications and ethical issues [6-8], identifying nurses' perceptions of using physical restraints on the hospitalized older adults is of particular importance [15, 16, 19-22]. This study aims to determine the psychometric properties of the Persian version of the Perceptions of Restraint Use Questionnaire (PRUQ).

This is a methodological study with a descriptive cross-sectional design that was conducted in 2019 in hospitals of Kermanshah city (West Iran). After obtaining ethical approval from the ethics committee of the University of Social Welfare and Rehabilitation Sciences (Code: IR.USWR.REC.1398.121) and obtaining permission from the developers of PRUQ, the questionnaire was translated into Persian according to the standard protocol of the World Health Organization and using the Forward-Backward method [31]. Then, face validity and content validity of the Persian PRUQ were evaluated qualitatively and quantitatively. To examine the construct validity of the Persian PRUQ, Exploratory Factor Analysis (EFA) and Confirmatory Factor (CFA) were used using a sample size of 330 patients. For EFA, the Persian PRUQ was completed by 170 nurses who were selected by a convenience sampling method based on the inclusion criteria (having at least one year of experience in adult intensive care, internal medicine and surgery, active involvement in decision-making to use physical restraint in hospitalized elderly, and consent to participate in the study). 
Kaiser-Meyer-Olkin (KMO) index and Bartlett’s test of sphericity were used for EFA. For CFA, 160 nurses were examined using multivariate techniques were used to evaluate the relationships. To fit the model, goodness of fit indices (χ2, χ2/df, RMSEA, GIF, NFI and CFI) were used. In order to examine the test-retest reliability using the Intraclass Correlation Coefficient (ICC), 30 nurses were selected and were asked to answer the questions two times with a two-week interval. The correlation between the scores of the two tests was determined using the ICC. Internal consistency was calculated using Cronbach's alpha for the whole questionnaire and for each item. Moreover, the reliability of the constructs was examined by examining the ratio of variance in the observed variables to the variance of latent variables in the CFA. Statistical analyzes were performed in SPSS v. 23 and AMOS v. 5 software.

Out of 330 nurses, 74.2% (n=245) were women and 63.9% (n=211) had a bachelor's degree in nursing. Their Mean±SD of age and work experience were 36.52±4.23 years and 8.22±4.86 years, respectively. In assessing qualitative face validity, one item (item 5) needed to be revised to remove ambiguity. In quantitative face validity evaluation, the effect size of all items was higher than 1.5. In qualitative content validity evaluation, the suggestions of geriatricians in terms of clarity and simplicity were applied, in addition to minor grammar corrections. In assessing quantitative content validity, a Content Validity Ratio (CVR) of 0.78 for the entire questionnaire and a scale CVR of 0.88 were obtained. Accordingly, all tool items were accepted. In construct validity evaluation, hidden factors were extracted based on the results of EFA. KMO test result for sample size adequacy was 0.672 which was acceptable and the result of Bartlett’s test of sphericity was statistically significant (P<0.001). 
In EFA using maximum likelihood method, Promax rotation and Gravel diagram, three factors "prevention of fall", "prevention of interruption or interference in treatment" and "providing a safe environment" were extracted. These three factors had eigenvalues of 3.58, 3.07 and 2.22, respectively, and explained 52.16% of the total variance of Persian PRUQ subscales. The results of CFA also provided a good estimate based on the fit indices of the model (χ2/df = 2.27, GFI = 0.83, CFI = 0.91, NFI = 0.92, RMSEA = 0.08). In the final model of the factor structure of the Persian PRUQ, the variables showed a high correlation with their respective factor. Moreover, there was a correlation between the measurement error in the items related to pulling out a catheter and a feeding tube (a6 and b6) and that of items related to prevention of breaking open sutures and removing a dressing (d6 and e6). The reliability of the Persian PRUQ was reported acceptable using Cronbach's alpha (0.82), structural reliability (>0.70) and ICC=0.86 (95%CI: 0.74-0.93, P <0.001).

The Persian version of PRUQ has acceptable validity and reliability. Due to the small number of items, using this questionnaire is easy and requires little time. Therefore, this questionnaire can be used to study and evaluate the perception of the use of physical restraint for geriatric care in Iran and compare the results with the studies conducted in other countries. This questionnaire can also be a good tool for evaluating interventions in the field of physical restraint of the elderly; by comparing the perception of using physical restraint before and after receiving various interventions, it is possible to realize the effectiveness of the interventions to a large extent. Identifying nurses' perceptions of using physical restraint in the elderly can be helpful in planning and policy-making to reduce the use of physical restraints and improve the quality of care for the hospitalized older people.

The Research Ethics Committee of the University of Social Welfare and Rehabilitation Sciences approved the study (Code: IR.USWR.REC.1398.121). All ethical principles are considered in this article. The participants were informed about the purpose of the research and its implementation stages. They were also assured about the confidentiality of their information and were free to leave the study whenever they wished, and if desired, the research results would be available to them.

This study was extracted from a PhD. dissertation of the first author at the Department of Nursing, University of Social Welfare and Rehabilitation Sciences, Tehran.

All authors equally contributed to preparing this article.

The authors declare no conflict of interest