Physical restraints among the elderly in the acute care setting: Prevalence,complications and its association with patients’ characteristics

Abstract

Aims:

This study aimed to identify the prevalence and circumstances of the use of physical restraints, complications associated with its use and its association with patients’ characteristics.

Methods:

This is a prospective observational study design. A convenience sample of elderly patients (aged 65 and above) who were admitted to an acute care hospital from February 2012 to August 2012 were tracked daily to monitor the use of physical restraints during hospitalization. Patient-related variables and type of physical restraint used were gathered from the medical records and the Multi-Dimensional Dementia Assessment Scale was completed through interviews with nurses who had cared for the patients. Independent t-test or Chi-square (χ²) tests were used to examine the effect of a single variable on the use of physical restraints. Multiple logistic regression was done to predict the patient-related variables to the likelihood of being placed on physical restraints.

Results:

Altogether, 998 patients accounting for 1048 admission episodes, were included in this study, of which 84 admission episodes involved the use of physical restraints (8%). Major reasons for restraints include ‘Behavioural/confused/violent’ (n= 55; 65%) and to prevent falls (n=53; 62%). Regression analysis indicated that male gender, behavioural changes, memory disturbance, impaired physical function, the use of anti-psychotic drugs and higher psychiatric workload predicted the likelihood of being put on physical restraints. The strongest predictor was memory disturbances with an odds ratio of 30.

Conclusion:

The prevalence of physical restraints used was higher than that in published literature for acute care hospitals. Preventive strategies would need to be targeted for the elderly, who suffered from memory disturbances upon admission.

Keywords

Geriatrics acute hospital nursing physical restrainer

Introduction

Physical restraint is commonly defined as:

‘Any device, material or equipment attached to or near a person’s body and which cannot be controlled or easily removed by the person and which deliberately prevents or is deliberately intended to prevent a person’s free body movement to a position of choice and/or a person’s normal access to their body’ (p. 2).¹

A common cited reason for using physical restraint was staff’s concern for the patient’s safety, in particular to the patient climbing out of bed and falling.² Nonetheless, more recent literature indicated that the use of physical restraints is associated with falls, pressure sores, depression, aggression and even death.³

The Joint Commission⁴ recommended that restraints should only be used during emergencies in which there is an imminent risk of a person physically harming himself or herself, staff or others and non-physical measures would not be effective. The Commission also advocates for organizations to collect data on the use of restraints in order to monitor and implement strategies to reduce the use of physical restraints.

In order to effectively reduce the use of physical restraints in the local context, nurses need to first understand why and when patients were placed on restraints, as well as how frequent and for how long. Evaluating the association between patients’ characteristics and use of physical restraints will also allow nurse managers to identify significant risk factors. Hence, this study aimed to identify the frequency, duration, type and reasons for the use of physical restraints in an acute teaching hospital in Singapore and its association with patients’ characteristics.

Specifically, the research questions were:

What was the prevalence of physical restraint use among elderly patients (aged ⩾65 years) over the six months period?

What type of physical restraints was used and their frequency and duration?

What were the reasons for using physical restraints?

What were the complications associated with the use of physical restraints?

What was the relationship between patients’ characteristics, demand on nursing hours and the use of physical restraints?

Literature review

Physical restraints have been commonly used in a bid to prevent patients from injuring themselves and/or others in the acute care settings.

Most of the studies conducted thus far were observational studies. The comparison of prevalence rate across studies was challenging due to disparate definitions of physical restraints and measuring units. In some studies, the use of bed rails was considered as a form of physical restraint while in other studies only the use of devices attached to a patient’s body, which prevent free body movements, were included. For example, Minnick et al.⁵ reported that the mean prevalence rate was 52.8 per 1000 patient days, ranging from 4.7 to 94 across 40 institutions in the USA. However, the authors only include cases whereby a physical restraint other than full side rails was used. In addition, the most common cited reason for physical restraints was to ‘prevent therapy disruption’ (74.9% of restrained days), followed by ‘confusion’ (25.4%) and ‘fall prevention’ (17.6%).

A more recent study by Barton-Gooden et al.⁶ reported a prevalence rate of as high as 75% in medical-surgical wards in a teaching hospital in Jamaica. However, they included partial and full bedrails forms of restraint and 70% of the cases were using partial and full bedrails and only 5% were using limb and trunk restrainers. The authors also reported that older patients were more likely to be restrained, although the results were not statistically significant. In another large scale study in Germany, researchers found that the prevalence of use of at least one physical restraint was on average 11.8% in four acute care hospitals (95% confidence intervals (CI) 7.8–15.7) and mostly involved the use of full bedrails.⁷

Most studies which look at the impact of physical restraints were done in mental health facilities and long-term care homes, with very few focusing on acute care facilities and the elderly. The only published local study on the prevalence of physical restraints reported that 23% of elderly nursing home residents were on physical restraints.⁸

Nonetheless, studies which were done demonstrated consistent results, i.e. patients who were on physical restraints had the worst outcomes. Patients who were restrained were more likely to fall, had a longer hospitalization stay, were at higher risk of nosocomial infections and pressure ulcers than patients who were not restrained.^9–11

Several studies have shown a relationship between patients’ characteristics and restraint use. Tinetti et al.¹² found that older age, disorientation, dependence on getting dressed and the use of antidepressants were independently associated with the likelihood of being physically restrained. Similarly, Hamers et al.¹³ reported that poor mobility, care dependency and fall risk in the opinion of nursing staff were predictors of restraint use. Unequivocally, the use of restraint was highly associated with patient’s mobility; the use of restraint was 11 times higher for those with poor mobility compared to a mobile person. However, these studies were all conducted in long-term care facilities; the results may not be generalizable to an acute care setting. In another study which investigated the influence of environmental and organizational variations and patient and staff characteristics on the prevalence of restraint use, it was found that the proportion of patients with impaired mobility function, the number of behavioural disturbances and nursing staff’s attitudes towards use of restraints were the strongest discriminators between restraint-free wards and wards that used restraints.¹⁴

Given the puacity of studies being conducted in the acute care setting and the lack of local published data, the present study provided a timely update on the use of physical restraints among the elderly in an acute care setting in Singapore. Results of the study will help to inform the design and implementation of strategies to reduce the use of physical restraints in the acute care setting.

Methods

This is a prospective observational study conducted in an acute care hospital in Singapore. A research assistant was trained by the study team (senior registered nurses) in the collection of data from medical records and the use of the Multi-Dimensional Dementia Assessment Scale (MDDAS). The research assistant gathered patient’s admission data (through the hospital’s admission information system) and patient’s demographics from the medical records. The patients who fit the inclusion criteria were monitored daily for any use of physical restraint until discharged. On the day of discharge, the research assistant went to the wards to complete the MDDAS, which was completed through interviews with the nurses who cared for the patient for at least two shifts. The MDDAS gathered information on patients’ motor function, vision, hearing, speech, activities of daily living functions, behavioural and psychiatric symptoms, use of psychoactive drugs as well as demand on nursing hours (based on nurses’ experienced physical and psychiatric workload). Nurses were asked to rate the physical and psychological workloads using a visual analogue scale with ratings from 1 (minimum) to 5 (maximum).¹⁵

For elderly patients who were put on physical restraints, data such as date and time of initiation, type of physical restraint used, reason for restraint use, duration of restraint, and complications resulting from use of physical restraints were gathered from the clinical documentation.

Sample

A convenience sample of elderly patients who were admitted to a large tertiary teaching hospital in Singapore were included in this study.

The inclusion criteria were:

patients aged 65 or above;

admitted to the hospital during the study period of February 2012 to August 2012 (six months).

The exclusion criteria were:

admitted to observation ward within the emergency department;

discharged within 24 hours.

Based on an estimated prevalence of 20%, it was calculated that at least 984 patients were required at a 95% CI and 0.05 width of interval.

Definition

The definition of physical restraints was adapted from the Joanna Briggs Institute, which stated that:

‘Physical restraint is any device, material or equipment attached to or near a person’s body and which cannot be controlled or easily removed by the person and which deliberately prevents or is deliberately intended to prevent a person’s free body movement to a position of choice and/or a person’s normal access to their body’ (p. 2).¹

In addition, this study does not consider partial or full bed rails as a form of physical restraint.

Ethical considerations

This study was approved by the SingHealth Centralised Institutional Review Board in Singapore (CIRB ref no: 2011/834/A). Permission to use the hospital administrative database was approved by the administrators and Chief Nurse of the hospital and the Information Technology department of the hospital. A waiver of informed consent was approved by the Ethics Committee.

Data analysis

All statistical analyses were performed using Statistical Package for Social Science software version 21. Patient’s demographic data, length of hospitalization and data for the type, duration and frequency of physical restraints use were analyzed using descriptive statistics. Independent t-test or χ² tests were used to examine the effect of a single variable such as patient’s age, physical and psychiatric workload on the use of physical restraints. Multiple logistic regression was done to predict the patient-related variables for the likelihood of being placed on physical restraints. Test of significance were two-tailed, with p value <0.05 considered significant.

Results

Sample characteristics

There were 998 unique patients accounting for 1048 admission episodes in this study. Out of these, 44 patients were admitted twice and three patients were admitted three times during the study period. Most subjects were females (n=559, 56.0%) and the average age was 77 years (SD=7.87, range 65–101). The average length of hospitalization was 10 days (SD=11.86, range 1–166).

A small proportion (n=133, 12.7%) of patients were prescribed with anti-psychotic drugs during their hospitalization. In most of the admission episodes, patients did not present with behavioural changes or memory disturbances (n=859, 82% and n=841, 80.2% respectively). The majority of patients had no paralysis (n=746, 71.2%), good vision (n=821, 78.4%) and good hearing (n=908, 86.7%) (Table 1).

Table 1.

Demographics of patients and patient-related variables.

Demographics			Mean (SD)	Range
Age			77 (7.87)	65–101
Length of stay			10 (11.86)	1–166
Physical workload			2.43 (1.22)	1–5
Psychiatric workload			1.84 (1.15)	1–5
			N	% of patients
Gender		Male	439	44
Gender		Female	559	56
Patient-related variables			N	% of hospitalization episodes
Prescribed with anti-psychotics		Yes	133	12.7
Prescribed with anti-psychotics		No	915	87.3
Behavioural changes (has the patient behaved and functioned this last week as he/she generally does?)		Yes	92	8.8
		No	859	82.0
		Missing data	97	9.3
Memory disturbances (does the patient have memory disturbances?)		Yes	118	11.3
		No	841	80.2
		Missing data	89	8.5
Paralysis	Impaired physical function	Unilateral paralysis	9	0.9
		Unilateral paralysis, partially retained strength	16	1.5
		Reduced strength in arm or leg	263	25.1
	No paralysis		746	71.2
	Missing data		14	1.3
Vision	Impaired vision	Blind	1	0.1
		Greatly reduced vision, often needs guidance	19	1.8
		Moderately reduced vision, no handicap	199	19.0
	Good vision		821	78.4
	Missing data		8	0.7
Hearing	Impaired hearing	Deaf	4	0.4
		Greatly reduced hearing, must speak directly into the ear	15	1.4
		Moderately reduced hearing, cannot manage normal conversational level	113	10.8
	Good hearing, responds to normal conversational level		908	86.7
	Missing data		8	0.7
On morning care, does patient require help of:		One staff member	618	58.9
		Two staff members	333	31.8
		No staff	92	8.8
		Missing data	5	0.5

In terms of workload, patients were graded an average of 2.43 for physical workload and 1.84 for psychiatric workload (from a scale of 1 to 5; 1 = minimal and 5 = maximal). The majority of patients needed only one staff member (n=618, 58.9%) to assist during morning care and very few were independent (n=92, 8.8%) (Table 1).

Prevalence of physical restraint use among the elderly

In 84 (8%) admission episodes, 83 patients were on physical restraints; 35% (n=29) were in a body vest, 35.7% (n=30) were put on a limb restrainer and 28.6% (n=24) were on both body and limb restrainers. All of the limb restrainers were only applied on the upper limbs and none on the lower limbs.

The duration of restraint use ranged between 0 (less than 1 day) and 70 days and the mean duration was 11 days (SD=12.0).

Reasons for applying and discontinuing physical restraints

The most common reasons for applying physical restraints were ‘to prevent falls’ (n=53, 62%) and to manage ‘confused or violent’ patients (n=55, 65%). Another commonly quoted reason was to prevent the patient from pulling out tubes and lines (n=50, 59%).

In most episodes (n=38, 45%), restraints were only removed when the patient was discharged. A change to hand mittens was identified as discontinuation of physical restraints in four cases (5%) (Table 2).

Table 2.

Use of physical restraints.

		N	% of hospitalization episodes
Being placed on physical restraints during admission	Yes	84	8
Being placed on physical restraints during admission	No	964	92
		Mean (SD)	Range
Duration of restraint (days)		11 (12)	0–70
		N	% of restraint episodes
Type of physical restraint	Body vest only	29	34.5
	Limb restraints only	30	35.7
	Body vest and limb restraints	24	28.6
	Body vest and limb restraints and seat belt	1	1.1
Reasons for physical restraints^a	Treatment or procedural	5	6
	Prevent pulling	50	59
	Prevent fall	53	62
	Behaviour confused or violent	55	65
Reasons for discontinuing physical restraints	Patient died	1	1.1
	Not required anymore	2	2.4
	Changed to mittens	4	4.8
	Cooperative (not restless, not pulling out tubes)	29	34.5
	Discharged	38	45.2
	No reason documented	10	11.9

Percentages add up to more than 100 as multiple reasons were given for some patients to be placed on physical restraints.

Complications associated with the use of physical restraints

Of the patients who were physically restrained, only nine (10.8%) fell during their hospitalization stay. However, the fall incidents were not associated with the use of physical restraints as all of these patients were put on physical restraints only after the fall incident.

Two (2.4%) developed nosocomial pressure ulcers during their hospitalization stay.

Relationship between patient’s characteristics and physical restraints use

In order to evaluate the relationships between patient’s characteristics and use of physical restraints, only details of patients’ first admission episode during the data collection period were included; hence, only data of 80 patients who were restrained on their first admission episode were included in the analysis.

Patients who were on physical restraints tended to be slightly older (mean=81, SD=8.56) than those who were not restrained (mean=77.3, SD= 7.7), t (996) =4.027, p=< 0.001 but the magnitude of effect was very small (η²=0.018).

In terms of reported physical workload and psychiatric workload, there was a statistically significant difference between patients who were on physical restraints vs. those who were not restrained. Physical workload was higher for patients who were restrained (mean=3.36, SD= 1.05) compared to those who were not restrained (mean=2.34, SD= 1.2), t (996) = −7.33, p=<0.001. Psychiatric workload was also higher for patients who were restrained (mean=2.91, SD=1.07) compared to those who were not restrained (mean=1.74, SD=1.09), t (996) = −9.16, p=<0.001. The magnitude of effect was moderate for both physical and psychiatric workload respectively (η²=0.06; η²=0.07).

In this study, male patients were more likely than females to be on physical restraints χ² (1, n=998) =11.3, p =0.0. Patients who had behavioural changes χ² (1, n=901) =132.6, p=0.0 and memory disturbance were also more likely to be on physical restraints χ² (1, n=909) =338.2, p=0.00. Patients who were prescribed with anti-psychotic drugs χ² (1, n=998) =17.2, p=0.00 and patients who had impaired physical function were also more likely to be on physical restraints χ² (1, n=984) =5.3, p=0.021. There were no significant associations between use of physical restraints, vision and hearing function.

Table 3 illustrates the demographic variables of patients who were and were not restrained and results of the univariate analysis.

Table 3.

Demographic variables of patients who were and were not restrained and results of univariate analysis.

Variable	Restrained (mean/SD)	Not restrained (mean/SD)	t-test (p value)	95% confidence interval of difference	η²
Age	81/8.56	77.3 /7.7	t=−4.027, 0.00	−5.4 to 1.87	0.018
Physical workload	3.36/1.05	2.34/1.2	t=−7.33, 0.00	−1.3 to 0.74	0.06
Psychiatric workload	2.91/1.07	1.74/1.09	t=−9.16, 0.00	−1.42 to 0.923	0.07
Variable		Restrained (%)	Not restrained (%)	χ² test (continuity correction value, p-value)	Φ coefficient
Gender	Male	50 (11.4)	389 (88.6)	11.3, 0.00	−0.113
Gender	Female	30 (5.4)	529 (94.6)
Behavioural changes	Yes	37 (42.5)	50 (57.5)	132.6, 0.00	−0.387
Behavioural changes	No	42 (5.3)	772 (94.7)
Memory disturbances	No	18 (2.3)	781 (97.7)	338.2, 0.00	0.62
Memory disturbances	Yes	61 (55.5)	49 (44.5)
Prescribed with anti-psychotics	No	57 (6.6)	810 (93.4)	17.2, 0.00	0.135
Prescribed with anti-psychotics	Yes	23 (17.6)	108 (82.4)
Paralysis	No paralysis	49 (6.8)	658 (93.2)	5.3, 0.021	0.075
Paralysis	Impaired physical function	32 (11.5)	246 (88.5)
Vision	Good vision	62 (7.9)	718 (92.1)	0.023, 0.88	0.008
Vision	Impaired vision	18 (8.6)	192 (91.4)
Hearing	Good hearing	68 (85.2)	800 (87.9)	0.339, 0.56	0.023
Hearing	Impaired hearing	12 (14.8)	110 (12.1)

Multiple logistic regression was performed to assess the impact of patient-related variables on the likelihood of them being physically restrained. The model contained variables which were significantly associated with the use of physical restraints in the univariate analysis.

The logistic model was statistically significant, χ² (7, n=889) =268.4 p=0.00, indicating that the model was able to differentiate between patients who were physically restrained and those who were not restrained. The model as a whole explained between 26.6% (Cox and Snell square) and 58.9% (Nagelkerke R²) of the variance in restraint status and correctly classified 93.9 % of cases.

As shown in Table 4, the variables which can predict the likelihood of being put on physical restraints were male gender, behavioural changes, memory disturbance, impaired physical function, the use of anti-psychotic drugs and psychiatric workload. The strongest predictor for physical restraints was memory disturbances with an odds ratio of 30.

Table 4.

Logistic regression predicting use of physical restraints.

Variables	B	SE	Wald	df	p-value	Odds	95% CI for odds ratio
						ratio	Lower	Upper
Age	0.030	0.021	2.088	1	0.148	1.03	0.989	1.073
Sex (male)	−1.330	0.352	13.950	1	0.000	0.264	0.132	0.531
Behaviour changes	−1.276	0.390	10.728	1	0.001	0.279	0.130	0.599
Memory disturbances	3.392	0.370	83.819	1	0.000	29.718	14.377	61.427
Physical function	1.030	0.359	8.242	1	0.004	2.802	1.387	5.660
Physical workload	0.043	0.165	0.066	1	0.797	1.043	0.755	1.443
Psychiatric workload	0.521	0.166	9.886	1	0.002	1.683	1.217	2.328
Psychotic drugs	1.033	0.406	6.466	1	0.011	2.811	1.267	6.234
Constant	−6.21	1.66	13.98	1	0.00	0.002

df: degrees of freedom.

Discussion

A total of 8% of the 1048 admission episodes involved elderly patients being put on physical restraints. Although the prevalence rate was lower than in a long-term care setting in Singapore⁸ and elsewhere, it was higher than the reported prevalence rate for acute care hospitals. To illustrate, prevalence rates ranged from 6% in nursing homes in Switzerland to over 31% in Canada,¹⁶ whereas, in acute care hospitals, a prevalence rate as low as 2.2% was reported.¹⁷ None-theless, it was difficult to benchmark our results against other studies due to the varying definitions of ‘physical restraints’ in the literature and differing methodologies used to collect data.¹⁸ Previous studies were based on point prevalence surveys⁵ over several days or self-reported data from nurses.¹⁹ However, in this study, patients were monitored daily throughout their hospitalization for the use of the physical restraints.

In an earlier study by Minnick et al.,²⁰ the most common type of restraint was wrist restraints, followed by vest or jacket restraint and the two most frequent reasons for the use of restraints were ‘prevention of disruption of therapy’ and ‘fall prevention’. This finding was congruent with the results of the current study, whereby the most common types of restraints used were body vest and upper limb restraints. Similarly, in this study, physical restraints were mostly used to manage confused or violent patients and to prevent falls. Although none of the patients in our study fell while being physically restrained, the use of physical restraints to prevent falls is frowned upon given the growing evidence that physical restraint may actually increase a patient’s risk of falling and injuring themselves.^11,21

Previous studies^14,22 concluded that the frequency of physical restraint use in nursing homes and group living was related to the behaviour of the residents, their cognitive and mobility statuses. Similarly, in our study, the use of physical restraints was significantly associated with patients’ characteristics and the most significant factor is patients with memory disturbances.

Implications for practice

The findings highlight the urgent need to devise and implement strategies to reduce the use of physical restraints in the acute care setting. Preventive strategies would need to be targeted upon admission to the hospital for the elderly who have memory disturbances. Individualized care and greater involvement and education of patients and their family members on care plans are required in order to provide a restraint-free care environment, especially in an acute care setting.

Nurses also need to be educated on the adverse effects of physical restraints on falls, effective strategies to prevent falls and be empowered (with resources and training) to implement them.

Limitations and recommendations for future research

The assessment of physical and psychiatric workload was done by different nurses in different wards, with varying years of work experience; hence, the accuracy of estimation of workload might be affected. It is recommended that future research involve the use of a small group of nurse assessors and establish inter-rater reliability on the assessment of the patient’s physical and psychiatric workload. In addition, patient-related variables may need to be interpreted with caution as the missing values are rather high (n=97, 9.3%; n=89, 8.5%) respectively in Table 1. There is a difference between missing values and data which is not applicable.

This study also did not measure the knowledge and attitudes of nurses towards the use of physical restraints as well as the culture of the different wards. These variables could have impacted the way nurses managed elderly patients with memory disturbance and/or behavioural change; therefore, future studies should consider studying the beliefs, knowledge and attitudes of nurses toward the use of physical restraints.

Conclusion

The prevalence of physical restraints used was higher than that in the published literature for acute care hospitals. Preventive strategies would need to be targeted for the elderly who suffer from memory disturbances upon admission.

Footnotes

Conflict of interest

The authors declared no conflict of interest in the conduct of this study.

Funding

This research is supported in funding by Singapore Ministry of Health Nursing Research Committee Grant (Grant number: MOHNRC FY 11-08).

References

Best practice. Physical Restraint – Part 1: Use in Acute and Residential Care Facilities Best Practice: Evidence Based Practice Information Sheets for Health Professionals, 2002; 6: 1–6, http://connect.jbiconnectplus.org/ViewSourceFile.aspx?0=4326 (2002, accessed 8 April 2015).

Frank

Hodgetts

Puxty

. Safety and efficacy of physical restraints for the elderly. Review of the evidence. Can Fam Physician 1996; 42: 2402–2409.

Hamers

JPH

Huizing

. Why do we use physical restraints in the elderly? Z Gerontol Geriat 2005; 38: 19–25.

Joint Commission. Implementation guide for the NQF endorsed nursing-sensitive care measure set. Jt Comm Oakbrook Terrace, http://www.jointcommission.org/assets/1/6/NSC%20Manual.pdf (2009, accessed 12 December 2013).

Minnick

Mion

Johnson

. Prevalence and variation of physical restraint use in acute care settings in the US. J Nurs Scholarsh 2007; 39: 30–37.

Barton-Gooden

Dawkins

Bennett

. Physical restraint usage at a teaching hospital: A pilot study. Clin Nurs Res 2013; June: 1–18.

Krüger

Mayer

Haastert

Meyer

. Use of physical restraints in acute hospitals in Germany: A multi-centre cross-sectional study. Int J Nurs Stud 2013; 50: 1599–1606.

Mamun

Lim

. Use of physical restraints in nursing homes: current practice in Singapore. Ann Acad Med 2005; 34: 158–162.

Rakhmatullina

Taub

Jacob

. Morbidity and mortality associated with the utilization of restraints: A review of literature. Psychiatr Q 2013; 84: 499–512.

10.

Castle

Engberg

. The health consequences of using physical restraints in nursing homes. Med Care 2009; 47: 1164–1173.

11.

Evans

Wood

Lambert

. Patient injury and physical restraint devices: A systematic review. J Adv Nurs 2003; 41: 274–282.

12.

Tinetti

Liu

Marottoli

. Mechanical restraint use among residents of skilled nursing facilities. Prevalence, patterns, and predictors. JAMA 2015; 265: 468–471.

13.

Hamers

Gulpers

Strik

. Use of physical restraints with cognitively impaired nursing home residents. J Adv Nurs 2004; 45: 246–251.

14.

Karlsson

Bucht

Eriksson

. Factors relating to the use of physical restraints in geriatric care settings. J Am Geriatr Soc 2001; 49(: 1722–1728.

15.

Karlsson

Bucht

Eriksson

. Physical restraints in geriatric care in Sweden: prevalence and patient characteristics. J Am Geriatr Soc 1996; 44: 1348–1354.

16.

Pasquier

N Du

Gilgen

Ikegami

. Use of physical restraints and antipsychotic medications in nursing homes: a cross-national study. Int J Geriatr Psychiatry 2013; 24: 1110–1118.

17.

Cosper

Morelock

Provine

. Please Release Me: restraint reduction initiative in a health care system. J Nurs Care Qual 2015; 30: 16–23.

18.

Kruger

Meyer

Hamers

. [Physical restraints in hospital. A systematic overview]. Zeitschrift fur Gerontologie und Geriatrie. 2010; 43: 291–296.

19.

Degan

Iannotta

Genova

. The use of physical restraints in an acute care hospital. Assist Inferm Ric 2004; 23: 68–75.

20.

Minnick

Mion

Leipzig

. Prevalence and patterns of physical restraint use in the acute care setting. J Nurs Adm 1998; 28: 19–24.

21.

Wilschek

Miller

Wetzel

. Restraint use, restraint orders, and the risk of falls in hospitalized patients. Am Geriatr Soc 2002; 50: 526–529.

22.

Huizing

Hamers

JPH

de Jonge

. Organisational determinants of the use of physical restraints: a multilevel approach. Soc Sci Med 2007; 65: 924–933.