Use of physical restraints in acute hospitals in Germany: A multi

Transcription

G Model
NS-2226; No. of Pages 8
International Journal of Nursing Studies xxx (2013) xxx–xxx
Contents lists available at SciVerse ScienceDirect
International Journal of Nursing Studies
journal homepage: www.elsevier.com/ijns
Use of physical restraints in acute hospitals in Germany:
A multi-centre cross-sectional study
Cäcilia Krüger a, Herbert Mayer a, Burkhard Haastert b, Gabriele Meyer a,c,*
a
Faculty of Health, School of Nursing Science, Witten/Herdecke University, Witten, Germany
mediStatistica, Neuenrade, Germany
c
Medical Faculty, Institute for Health and Nursing Science, Martin-Luther-University Halle-Wittenberg, Germany
b
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 18 July 2012
Received in revised form 6 May 2013
Accepted 7 May 2013
Background: Physical restraints are contrary to patients’ autonomy and freedom. Their
justification for controlling psychomotor agitation and risk of falling is being questioned
more and more often. Physical restraints are associated with many negative outcomes.
The German law is explicit, allowing physical restraints in nursing only as an exception.
Data on the use of physical restraints in acute hospitals in Germany are sparse.
Objectives: To investigate the prevalence of physical restraints and characteristics
associated with physical restraint use in acute hospitals.
Design: Cross-sectional study.
Participants and setting: 1276 patients (mean age 65 years, 45% women, 50% surgical) on
61 wards (n = 47 general; n = 14 intensive care) in four acute care hospitals in North RhineWestphalia, Germany.
Methods: One investigator visited each hospital ward at three randomly allocated time
slots on randomly selected days within a period of three months. A total of 3434 direct
observations on physical restraint status were collected. The study period lasted from
October 2008 to March 2009. For analysis, one time slot per patient room was randomly
chosen in order to avoid repeated analysis of the same patient.
Results: The prevalence of patients with at least one physical restraint was 11.8% (95%CI
7.8–15.7). The measures used most often were full bed rails (9.8%, 95%CI 6.5–13.1). There
was pronounced prevalence variation throughout the wards (general wards: 0.0–31.3%;
intensive care: 0.0–90.0%). The prevalence of physical restraints between hospitals ranged
from 6.2 to 16.6%, the overall association with hospital was non-significant. Multivariate
regression analysis revealed statistically significant characteristics for physical restraint
use: age 80–99 years versus 18–54 years (adjusted odds ratio 4.34, 95%CI 2.18–8.64),
feeding tube (2.70, 1.40–5.22), indwelling urinary catheter (6.52, 3.75–11.34), and staying
in intensive care unit (3.39, 1.29–8.92). Sharing a multi-bed room (0.55, 0.35–0.89) and in
situ central venous line were inversely associated (0.44, 0.19–0.98).
Conclusions: Physical restraints are apparently standard care in German acute hospitals.
However, variation between wards indicates that hospital care with only few physical
restraints is feasible. Respecting patients’ dignity and integrity warrants intervention
programmes aimed at decreasing practice variation towards a general reduction of
physical restraints in acute hospitals in Germany.
ß 2013 Elsevier Ltd. All rights reserved.
Keywords:
Epidemiology
Hospitals
Nursing
Restraint
Physical
* Corresponding author at: Medical Faculty, Institute for Health and Nursing Science, Martin-Luther-University Halle-Wittenberg, Germany,
Tel.: +49 345 557 4498; fax: +49 345 557 4471.
E-mail address: Gabriele.Meyer@medizin.uni-halle.de (G. Meyer).
0020-7489/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijnurstu.2013.05.005
Please cite this article in press as: Krüger, C., et al., Use of physical restraints in acute hospitals in Germany: A multi-centre
cross-sectional study. Int. J. Nurs. Stud. (2013), http://dx.doi.org/10.1016/j.ijnurstu.2013.05.005
G Model
2
C. Krüger et al. / International Journal of Nursing Studies xxx (2013) xxx–xxx
What is already known about the topic?
Former international epidemiological studies indicate
substantial use of physical restraints in the acute hospital
setting.
Data on the use of physical restraints in acute hospitals in
Germany are sparse.
What this paper adds
Physical restraints are applied to approximately 12 out of
100 patients in German acute care hospitals.
Restrictive bedrails are the physical restraint most often
used.
Variation of prevalence between wards is pronounced,
but indicates that hospital care with only few physical
restraints is feasible.
1. Background
Physical restraints are common practice in the acute
care setting (Demir, 2007; Martin and Mathisen, 2005;
Minnick et al., 2007; Heinze et al., 2012). They are used to
keep patients in bed, to prevent accidental falls or patients’
interference with therapy (Agens, 2010; Benbenbishty
et al., 2010; Lane and Harrington, 2011), even though many
negative outcomes are known such as decreasing mobility,
increased risk of pressure sores or psychological problems
(Agens, 2010). A restraint-free nursing care environment is
called for as a high standard of care (RNAO, 2012).
Physical restraints violate patients’ autonomy and
freedom and their right to take risks (Sokol, 2010).
Although not much research on patients’ perception of
physical restraints has been done, existing evidence
indicates that patients experience negative psychological
impact and negative feelings such as anger and fear, and
often they do not know why they are being restrained
(Bower and McCullough, 2000; Strout, 2010). Nursing staff
report feelings of guilt and embarrassment while using
physical restraints (Janelli et al., 2006), but mistakenly
overestimate the perceived benefits (Martin, 2002).
Bed rails, belts and chairs with a table are the measures
most often used (Minnick et al., 2007). A recently
published secondary data analysis of perceptions by
nurses on general wards in 15 German hospitals revealed
a prevalence of 9.3% physical restraints in a sample of 2827
patients (Heinze et al., 2012). Prevalence of physical
restraints in acute hospitals reported in international
publications ranges between 3% and 25% (Evans et al.,
2002; Agens, 2010). The variance can be explained by
different definitions of physical restraints, different data
collection techniques, the characteristics of the care
settings and the case mix (Krüger et al., 2010). Legal
regulations and tradition also determine the likelihood of
physical restraint use (Bower et al., 2003). The German law
is explicit, allowing physical restraints in acute hospitals
only for circumscribed exceptions with judicial authorisation for people who could not consent, or for a short period
of time in the case of an emergency (Fogel and Steinert,
2012).
Data on the use of physical restraints in acute hospitals
in Germany are sparse. High quality epidemiological data
are the basis for the future development of interventions in
order to achieve nursing care with a minimum of physical
restraints. Therefore, we performed a cross-sectional study
on the prevalence of physical restraints in acute care
hospitals and investigated associations with restraint use.
2. Methods
2.1. Recruitment
Recruitment took place in July 2008 in North RhineWestphalia, Germany. We invited a convenience sample of
four large maximum-care hospitals to take part in our
study. The hospitals were known from former research
projects and belong to a collaborative network, initiated by
Witten/Herdecke University, aimed at improving knowledge transfer from research to practice. None of the
hospitals had previously given priority to the topic of
physical restraints. All agreed to participate. Each hospital
nominated a nurse as contact person for the researchers
and all study-related issues on-site. The selection of
participating wards was made by the contact nurse of
the respective hospital, who was requested to select nearly
the same number of surgical and internal medicine units as
well as intensive care units. Paediatric and psychiatric
wards as well as accident and emergency departments
were excluded.
The study population consisted of patients aged over 18
years, who were cared for on the participating wards at the
time of data collection for the cross-sectional study.
2.2. Definition of physical restraints
According to an established definition, physical
restraints were 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’’
(Evans et al., 2002). Physical restraints included restrictive
bed rails, i.e. bilateral full bed rails or a full bed rail on one
side of the bed with the other side next to the wall, belts in
beds or chairs (wrist, elbow, ankle, abdominal), geriatric
chairs with fixed tables and other measures. Other
measures included tipping chairs, blankets or sheets,
and manipulation of furniture, all of which were considered as eligible. These devices and techniques have been
observed in nursing settings during our own previous data
collections on physical restraints (Meyer et al., 2009;
Köpke et al., 2012).
2.3. Data collection
Nursing staff on the wards were generally informed
about the upcoming data collection, but not about the date
and time. The contact nurse of each hospital was the only
person who was fully informed, but who was also
requested to conceal the dates. In preparation of the data
G Model
3
The prevalence data collection sheet contained the
options of physical restraints as defined in Section 2.2 and
patient characteristics as displayed in Table 1. Patient
characteristics were collected from observation and/or by
nursing staff who assessed patient records. The data
collection sheet was piloted for its feasibility and acceptability on two wards of a hospital not contributing to the
main study.
The investigator entered the patients’ rooms on her
own. The company of a staff nurse was not mandatory, but
optional, and generally took place in intensive care units.
For ethical reasons, the investigator was not allowed to lift
the blanket or actively search for physical restraints. Thus,
only apparent physical restraints could be documented. If
unsure whether a physical restraint was in place, a staff
nurse was asked. Only patients present in their rooms or on
the ward were included in the data collection.
collection, the contact nurse distributed laminated information sheets for patients and next of kin at the
participating wards to explain the upcoming in-house
data collection. The sheets were positioned so that they
were easily visible, or even fixed on the tables in the
patients’ rooms.
The data collection period throughout all the hospitals
lasted six months (from October 2008 until March 2009).
Data were collected by one external investigator (CK) who
was temporarily employed by the hospitals without
getting a salary.
Each participating hospital ward was visited three
times: in the morning (9:30 to 11:30 a.m.), in the afternoon
(4:00 to 6:00 p.m.), and in the evening (7:30 to 9:30 p.m.).
The time slots were determined not for feasibility only, but
also for the protection of patients’ private sphere by trying
to avoid disturbing them when sleeping or when using the
bathroom, for instance. In order to avoid underestimation
of physical restraint use due to projectable visits, the three
visits per ward were randomly allocated to randomly
selected days during a period of three months maximum
per hospital. Random procedure was carried out by a
researcher who was not engaged in the study and who
picked labelled and sealed notes out of a container.
2.4. Sample size
The outcome of the study was defined as patients with
at least one physical restraint documented at the time of
observation. Assuming a literature-based prevalence of
10% patients with restraint use (Krüger et al., 2010) and
Table 1
Characteristics of the randomly selected population.
Characteristic
Total random sample
(n = 1276)a
Patients with physical restraints on
general wards (n = 82)
Patients with physical restraints on
intensive care units (n = 68)
Women
Mean SD age, years
Age group (age classes around
quartiles of 54, 68, 78), years
18–54
55–69
70–79
80–99
Assisted breathingb
Line or drainagec
Feeding tube
Indwelling urinary catheter
Central venous line
Peripheral line
Arterial line
Surgical patients
Non-surgical patients
Assessed for physical restraint use on
weekdays (Monday to Friday)
Assessed for physical restraint use in the
Morning
Afternoon
Evening
Admitted to a ward size of
20 beds
>20 beds
Admitted to a ward size of
30 beds
>30 beds
Placed in
Multi-bed room
Single-bed room
Single occupancy in multi-bed room
577 (45)
65 18
49 (60)
79 13
24 (35)
62 20
(26)
(29)
(25)
(21)
(24)
4
10
18
50
32
(5)
(12)
(22)
(61)
(39)
19
17
20
12
59
(28)
(25)
(29)
(18)
(87)
126 (10)
328 (26)
241(19)
530 (42)
119 (9)
634 (50)
642 (50)
911 (71)
15
49
8
52
25
57
68
(18)
(60)
(10)
(63)
44
64
54
12
48
41
27
45
(65)
(94)
(79)
(18)
(71)
(60)
(40)
(66)
361 (28)
405 (32)
510 (40)
31 (38)
17 (21)
34 (41)
18 (26)
24 (35)
26 (38)
265 (21)
1011 (79)
9 (11)
73 (89)
53 (78)
15 (22)
578 (45)
698 (55)
29 (35)
53 (65)
64 (94)
4 (6)
981 (77)
146 (11)
149 (12)
57 (69)
8 (10)
17 (21)
32 (47)
25 (37)
11 (16)
329
369
313
265
307
(30)
(70)
(83)
Note: Values are numbers (percentage) unless stated otherwise.
a
Selected population (1 daytime per room, randomly selected) out of the total number of observations (n = 3434).
b
Assisted breathing means both artificial ventilation or oxygenation.
c
Multiple answers possible; Missing values (n = 13).
G Model
4
taking into account 20% of unoccupied beds, we calculated
a number of at least 1200 observations in four acute care
hospitals in order to estimate a 95% confidence interval (CI)
of the prevalence with a precision of 2% (i.e. 95%CI 8–12%).
The calculation was not adjusted for cluster correlation.
Additional cluster adjustment using the methods of Donner
and Klar (2000) and assuming a design factor of 5 (as
observed in the data, Table 2) would result in a larger 95%CI of
6–14% based on the same sample size.
2.5. Statistical methods
Since each hospital ward was visited three times – in
the morning, in the afternoon and in the evening – with
time slots in random order and on randomly selected
days within a period of three months maximum, it
cannot be ruled out that some patients were included in
the data collection more than once. Therefore, for
statistical analysis one time slot per room was
randomly chosen in order to avoid repeated analysis
of the same patient. Selection of one time slot per ward
would have been even more restrictive. However, we
assumed that the probability of a patient moving to
another room within the same ward and therefore
being observed twice during our data collection to be
very low.
Baseline characteristics of the randomly selected
population are described as mean values standard
deviations and as numbers and percentages. The outcome
analysis is adjusted for cluster correlation with hospital
wards as cluster units. All the patients in a room are
documented at a fixed time. Therefore, we assumed that the
documented patients are all different individuals and no
further adjustment for repeated measurement has to be
considered. Assuming that cluster correlation is related to the
ward level and not to the patient room level, no further
cluster adjustment for patients’ rooms was taken into
account.
Prevalence of physical restraints was estimated including cluster adjusted 95% confidence intervals using cluster
size weights as described by Donner and Klar (2000). The
cluster correlation was estimated by the corresponding
intracluster correlation coefficient.
Associations of characteristics of patients and institutions with the use of physical restraints were investigated
by multivariate logistic regression analysis adjusted for
cluster correlation using generalised linear mixed models
(Brown and Prescott, 2006). The dependent variable was
the use of physical restraints. Selected characteristics of
patients and institutions were investigated as fixed effects.
Hospital (n = 4) in particular are fixed effects. Clusters
(wards) are random effects. The Satterthwaite method was
used to calculate denominator degrees of freedom for the
tests of fixed effects.
Variable selection of the fixed effect was performed in
two steps. First, univariate models were built-in. All
variables with a p < 0.20 (testing the corresponding odds
ratio) were selected for the final multiple model. Performing stepwise or backward variable selection in multiple
regression cut-points for p-values of 0.10–0.15 is common
(Le, 2003). The higher cut-point of 0.20 should help to
avoid undetected associations due to confounding factors.
Also for this reason, an additional model was integrated
using all variables mentioned above (data not shown).
Age was analysed in the models using four classes
corresponding to rounded quartiles (18–54, 55–69, 70–79,
80–99 years). Ward size was dichotomized around the
mean [mean = 30; median = 33 (general ward (mean = 32;
median = 34); intensive care units (mean = 20; median = 16)]. Interpretation of the results was much more
difficult from a non-mathematical point of view when age
and ward size were analysed as continuous variables.
Statistical calculations were performed by SAS version
9.3 on Windows 7 64. Generalised mixed models were
built-in using SAS PROC GLIMMIX. Further data analyses
were performed by PASW Statistics (Version 18 and 20).
Table 2
Prevalence of physical restraints in the randomly selected population (n = 1276).
Patients with at least one physical restraintc
Patients with observed application of
Restrictive bedrails
Unilateral wrist restraintd
Bilateral wrist restraint
Waist belt in bedd
Chair with fixed tabled
Cluster adjusted prevalence, % (95%
confidence interval)
ICCCa
DFb
11.8 (7.8–15.7)
0.1751
5.064
0.1316
0.0343
0.2778
0.0380
0.0083
4.054
1.797
7.447
1.881
1.193
9.8
0.5
2.5
0.1
0.4
(6.5–13.1)
(0.0–1.1)
(0.2–4.9)
(0.0–0.3)
(0.0–0.8)
a
ICCC = Intracluster correlation coefficient.
DF = Design factor.
c
General wards (n = 1089 patients): cluster adjusted prevalence of patients with at least one physical restraint 7.5% [(95%CI 5.3–9.7), ICCC = 0.0412,
DF = 1.998], restrictive bedrails 7.1% [(95%CI 4.9–9.3), ICCC = 0.0436, DF = 2.056], unilateral wrist restraint 0.2% [n = 2, (95%CI 0.0–0.4), ICCC = 0.0000,
DF = 1.000], bilateral wrist restraint 0.2% [n = 2, (95%CI 0.0–0.4), ICCC = 0.0005, DF = 1.011], waist belt in bed 0.1% [n = 1, (95%CI 0.0–0.3), ICCC = 0.0429,
DF = 2.039], chair with fixed table 0.3% [n = 3, (95%CI 0.0–0.6), ICCC = 0.0000, DF = 1.000]; Intensive care units (n = 187 patients): cluster adjusted
prevalence of patients with at least one physical restraint 36.4% [(95%CI 23.2–49.6), ICCC = 0.1501, DF = 3.592], restrictive bedrails 25.7% [(95%CI 13.1–
38.3), ICCC = 0.1719, DF = 3.969], unilateral wrist restraint 2.7% [(95%CI 0.1–5.2), ICCC = 0.0116, DF = 1.200], bilateral wrist restraint 16.0% [(95%CI 5.4–
26.7), ICCC = 0.1739, DF = 4.005], chair with fixed table 1.1% [n = 2, (95%CI 0.0–2.8), ICCC = 0.0166, DF = 1.287]. No waist belt in bed was observed
at intensive care units.
d
Due to low prevalence of these physical restraints (less than 10 cases) the asymptotical estimation of cluster adjusted 95%CI, ICCC and DF should be
interpreted with caution.
b
G Model
2.6. Ethical approval
The protocol was approved by the ethics committee of
the School of Nursing Science at Witten/Herdecke University (June 2008). A waiver of consent from participating
patients was obtained, a method that has been successfully
applied in other own studies (Meyer et al., 2009; Köpke
et al., 2012). Additional approval was obtained by the local
ethical committees of the participating hospitals.
Due to data protection regulations, we were not
allowed to collect information about diagnoses, medication or cognitive status from patients’ records. All patientrelated data were de-identified.
3. Results
Sixty-one wards were visited, 47 were general wards
and 14 intensive care units. Wards caring only for surgical
patients (n = 26) had a mean cluster size of 21.7 9.1,
wards caring only for non-surgical patients (n = 30) of
19.6 7.5 and wards caring for surgical and non-surgical
patients (n = 5) of 25.0 9.4 (separated for different patients:
14.0 4.3 surgical patients, 11.0 6.4 non-surgical patients).
A total of 3434 observations were performed and made
available for outcome analysis (general wards: n = 2909;
intensive care units: n = 525). Half of the observations
(n = 1775, 52%) were performed on non-surgical wards and
half on surgical wards (n = 1659, 48%).
Random selection of one time slot per room revealed a
study population of 1276 patients. Mean age of the
randomly selected patient population was 65 18 years,
45% were women; 26% had an indwelling urinary catheter
and 42% an in situ peripheral line. Table 1 displays the sociodemographic and clinical characteristics of the randomly
selected population.
The prevalence of patients with at least one physical
restraint was 11.8% (95% confidence interval (CI) 7.8–
15.7%). The measures most often observed were bed rails
(9.8%, 95%CI 6.5–13.1), followed by bilateral wrist restraint
(2.5%, 95%CI 0.2–4.9). Table 2 presents the data on the
prevalence of the observed physical restraints.
Centre variation across wards was pronounced; prevalence ranged from 0.0% to 31.3% on general wards and
0.0% to 90.0% in intensive care units. The prevalence of
physical restraints between hospitals varied between 6.2%
and 16.6%, the overall association with hospital was nonsignificant (multivariate logistic model: p = 0.065, Table 3).
A set of 15 baseline variables were evaluated in
univariate models. The variables ‘female’ and ‘assessed
for physical restraint use in the afternoon’ were excluded
for the multivariate model because the related p-value was
above 0.20 (p = 0.46 and p = 0.34).
The results of the cluster adjusted univariate and
multivariate regression analyses are displayed in Table 3.
Multivariate regression analysis revealed statistically
significant positive associations with physical restraint
use for age 80–99 years versus 18–54 years (adjusted odds
ratio: AOR = 4.34, 95%CI = 2.18–8.64), in situ feeding tube
(AOR = 2.70, 95%CI = 1.40–5.22), indwelling urinary catheter (AOR = 6.52, 95%CI = 3.75–11.34) and staying in an
intensive care unit (AOR = 3.39, 95%CI = 1.29–8.92). The
5
hospital itself was not statistically significantly associated
with the use of physical restraints. Inverse associations
were found for sharing a multi-bed room (AOR = 0.55,
95%CI = 0.35–0.89) and in situ central venous line
(AOR = 0.44, 95%CI = 0.19–0.98). The results from an
additional model including all variables from univariate
analyses were very similar (data not shown).
4. Discussion
This cross-sectional study is the first in Germany that
determines the prevalence of physical restraints in a large
group of patients from four maximum-care hospitals by
direct observation, which is undoubtedly the most valid
and reliable method (Laurin et al., 2004). Although the
hospitals were personally known from an already existing
collaborative network, they do not represent a special
selection. Physical restraints were not a topic the hospitals
had focused on previous to our study.
We found that 12 out of 100 patients were subject to at
least one physical restraint, predominately restrictive bed
rails. The prevalence found in our study is slightly higher as
the prevalence of 9.3% reported by a recent secondary data
analysis of a questionnaire-based survey (Heinze et al.,
2012). Here, 2827 patients on general wards of 15 hospitals
in Germany were analysed, where – contrary to our study –
intensive care units were excluded from data collection.
We found a lower prevalence of 7.5% (95%CI 5.3–9.7) on
general wards, which might be due to different methodological procedures and characteristics of participating
hospitals.
Comparison to other studies is hardly feasible due to
different definitions of physical restraints, varying data
collection methods and the different settings and populations investigated (Krüger et al., 2010). In the USA, Minnick
et al. (1998) used comparable data collection methods, but
excluded bed rails. Data were presented as observation
rate of physical restraint use. The results indicated 5.8%
physical restraints out of 49,000 patient observations
throughout general wards and intensive care units. The
average use rate in intensive care units was much higher at
24.3%. Differences between wards were pronounced.
Our study confirms this variation of the prevalence
between wards of the same discipline. The reason is
unclear. The finding corresponds not only with the early
study by Minnick et al. (1998), but also with further
research in the acute hospital setting (Healey et al., 2009;
Minnick et al., 2007) and other nursing settings (Meyer
et al., 2009; Feng et al., 2009).
We also found that staying in an intensive care unit
implies a higher likelihood of physical restraint use. A
reason often given for more frequent restraint use in
intensive care units is the intended protection of patients
from self-initiated removal of indwelling devices (Benbenbishty et al., 2010; Turgay et al., 2009). In a recent study
(Benbenbishty et al., 2010) the prevalence of physical
restraints across all 566 patients from 34 intensive care
units in nine European countries was 39%. This figure is
quite comparable to our findings (Table 2: cluster adjusted
prevalence 36.4%, 95%CI 23.2–49.6), although our intensive care unit sample comprised only 187 patients. The
G Model
6
Table 3
Results of the cluster adjusted logistic regression analyses in the randomly selected population (n = 1276).
Characteristic
Univariate analysis
p-value
AORb (95%CI)
Female sex
Age (ref.: 18–54)
55–69
70–79
80–99
Assisted breathing
Feeding tube
Indwelling urinary catheter
Central venous line
Peripheral line
Arterial line
Intensive care unit patient
Surgical patient
Assessed for physical restraint use at the weekend
Assessed for physical restraint use in the afternoon
(ref.: morning and evening)
Ward size > 30 beds
Multi-bed room (ref.: 1-bed room and single
occupancy in multi-bed room)
Hospital (ref.: A)
B
C
D
a
b
c
p-value
AORb (95%CI)
1.15 (0.79–1.68)
1.24
2.31
5.96
5.47
8.02
12.36
2.39
1.33
3.69
8.24
0.62
0.71
0.81
Multivariate analysisa
(0.66–2.36)
(1.24–4.31)
(3.25–10.96)
(3.51–8.53)
(4.68–13.75)a
(7.81–19.56)a
(1.45–4.05)a
(0.89–1.99)a
(1.98–6.88)a
(4.39–15.45)
(0.33–1.19)
(0.44–1.13)
(0.53–1.24)
0.50 (0.24–1.03)
0.42 (0.28–0.63)
0.92 (0.26–3.22)
0.36 (0.10–1.24)
1.11 (0.30–4.12)
0.456
<0.001c
0.505
0.009
<0.001
<0.001
<0.001
<0.001
0.001
0.162
<0.001
<0.001
0.149
0.148
0.340
–
0.060
<0.001
1.11 (0.57–2.18)
0.55 (0.35–0.89)
0.750
0.015
1.21 (0.47–3.13)
0.49 (0.19–1.28)
1.25 (0.44–3.58)
0.065c
0.678
0.140
0.664
0.083c
0.897
0.104
0.876
1.10
1.86
4.34
1.67
2.70
6.52
0.44
1.20
0.79
3.39
0.67
0.78
–
(0.55–2.20)
(0.95–3.66)
(2.18–8.64)
(0.93–3.00)
(1.40–5.22)
(3.75–11.34)
(0.19–0.98)
(0.71–2.02)
(0.34–1.83)
(1.29–8.92)
(0.38–1.20)
(0.46–1.31)
–
<0.001c
0.791
0.072
<0.001
0.088
0.003
<0.001
0.045
0.500
0.584
0.014
0.175
0.342
–
Missing values (n = 13).
AOR = Adjusted odds ratio.
Overall type 3 test.
European comparative study (Benbenbishty et al., 2010)
revealed a country variation between 0.0% and 100%.
However, the result must be interpreted with caution,
since the country-related sample sizes differed from 15 to
319 patients.
Attitudes and beliefs of nursing staff seem to be decisive
factors for restraint use (Werner and Mendelsson, 2001;
Hamers and Huizing, 2005; Goethals et al., 2012). A recent
synthesis of qualitative evidence (Goethals et al., 2012)
indicates the complexity of the decision-making process
on appropriate use of physical restraints, which is
influenced by context and nurse-related factors. As also
pointed out in earlier work (e.g. Karlsson et al., 1998;
Meyers et al., 2001), the priority is on safety, giving higher
preference to physical restraint use rather than to
autonomy and freedom from restraints accompanied by
the risk of a patient falling or removing medical devices
(Goethals et al., 2012).
Legal regulation has certainly an impact on physical
restraint use in nursing. For instance, the introduction of
the US American Omnibus Budget Reconciliation Act in
1987 resulted in a relevant decrease of physical restraints
in nursing homes (Castle and Mor, 1998). However, even
strict legal regulations do not appear to adequately protect
nursing care recipients from physical restraints. German
law clearly directs that patients in acute nursing care
settings have free body movement. Physical restraints
should always be the last resort after all other alternatives
have been tested. Use of physical restraints in German
acute hospitals requires either the patient’s written
consent or judicial authorisation if the person is not able
to consent. Physical restraints are permitted as an
exception in the case of an emergency, if contractually
incapable, aggressive or violent persons show dangerous
behaviour that could not be otherwise controlled, or for
unconscious and sedated patients at intensive care units
who should be prevented from self-inflicted activities
(Fogel and Steinert, 2012). In our present study, we were
not able to collect information on judicial authorisation
and other justifications of physical restraints, since the
participating hospitals denied access to the data. It is
unknown whether acute care hospitals in Germany adhere
to the legal requirements of physical restraint use. With
regard to German nursing homes we know from our own
recent study (Köpke et al., 2012) that nowadays the
majority of physical restraints are authorised. However,
this finding could not be transferred to acute hospital care.
Physical restraints in German nursing homes and especially the legal issues have been a matter of debate for a
long time, quite contrary to the acute hospital setting.
Our regression analysis indicates the already known
positive associations between the use of physical restraints
and high age, indwelling catheter, and in situ feeding tube,
respectively (Heinze et al., 2012; Hamers and Huizing,
2005; Demir, 2007; Minnick et al., 2007). We also add some
new insights to the current knowledge on the use of
physical restraints in acute hospitals. The in situ central
venous line was inversely associated with the use of
physical restraints, a surprising result without self-evident
explanation. Sharing a multi-bed room was also inversely
associated with physical restraints. Again, the rationale
remains unclear. It may be that patients in single rooms
differ from patients in multi-bed rooms. Another explanation could be that nurses expect relatively healthy
roommates to control the behaviour of patients with an
increased risk of physical restraint use.
G Model
We could not find an association between gender and
the use of physical restraints as indicated by former studies
(Minnick et al., 1998, 2007).
Our study has its strong points: a large sample size was
investigated, both general wards and intensive care units
were included, the analysis took cluster correlation into
account and nursing staff were not aware of the date and
time of the visits. We can certainly not rule out that the
contact nurses informed the nursing staff in spite of
promising not to do so. The information sheets for patients
and next of kin distributed on the wards might have
influenced the frequency of physical restraints. However,
both potential limitations would have led to an underestimation of physical restraint use. Our study has other
limitations. We were not allowed to collect information on
diagnosis, medication or cognitive status from patients’
records because of data protection issues. The selection of
participating wards was made by the contact nurse at the
respective hospital. We could therefore not completely
rule out selection bias due to a non-representative sample
of wards.
Our study did not explore reasons for the use of physical
restraints because nurses are very likely to come up with
socially desired answers. Former studies (e.g. Hamers and
Huizing, 2005; Minnick et al., 1998, 2007) found fall
prevention, disruption of therapy or keeping patients from
wandering as the most important reasons for using
physical restraints.
Future research should reflect variation between wards.
For the development of interventional programmes aimed
at reducing physical restraints in acute hospitals, careful
exploration of practices used by centres with low
prevalence of physical restraints seems worthwhile.
International guidelines suggest alternatives for physical
restraints in acute care settings (Park and Tang, 2007; Bray
et al., 2004; Maccioli et al., 2003; RNAO, 2012) and all agree
that physical restraints should be used only as the last
resort after detailed assessment of the patient’s status, the
situation and the environment. The efforts should be
embedded in an extensive physical restraint reduction
approach comprising administrative support, interdisciplinary collaboration and staff education. This is a clear
challenge for all acute hospitals, not only in Germany,
because of the increasing number of admissions of patients
with dementia, reduced patients’ length of hospital stay,
and reduced nursing staff level (Rechel et al., 2009).
Acknowledgements
We thank the nurses and the participating patients of
the four hospitals in North Rhine-Westphalia. We also
thank Antonie Haut, MScN, Pia von Lützau, MScN, and
Britta Blotenberg, BScN, University of Witten/Herdecke, for
random selection of visits and data entry.
The study was conducted without external financial
support.
Conflict of interest: None declared.
Funding: None.
7
Ethical approval: The protocol was approved by the local
ethical committee of the Department of Nursing Science at
Witten/Herdecke University (available on request) and the
local ethical committees of the participating hospitals.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.ijnurstu.2013.05.005.
References
Agens, J.E., 2010. Chemical and physical restraint use in older person.
British Journal for Medical Practitioners 3 (1) 302.
Benbenbishty, J., Adam, S., Endacott, R., 2010. Physical restraint use in
intensive care units across Europe: the PRICE study. Intensive and
Critical Care Nursing 26 (5) 241–245.
Bower, F.L., McCullough, C.S., 2000. Restraint use in acute care settings:
can it be reduced? The Journal of Nursing Administration 30 (12)
592–598.
Bower, F.L., McCoullough, C.S., Timmons, M.E., 2003. A synthesis of what
we know about the use of physical restraints and seclusion with
patients in psychiatric and acute care settings: 2003 update. The
Online Journal of Knowledge Synthesis for Nursing 10 (April) 1.
Bray, K., Hill, K., Robson, W., Leaver, G., Walker, N., O’Leary, M., Delaney, T.,
Walsh, D., Gager, M., Waterhouse, C., British Association of Critical
Care Nurses, 2004. British Association of Critical Care Nurses position
statement on the use of restraint in adult critical care units. Nursing in
Critical Care 9 (5) 199–212.
Brown, H., Prescott, R., 2006. Applied Mixed Models in Medicine. Statistics
in Practice, 2nd ed. John Wiley & Sons Ltd, The Atrium, Southern Gate,
Chichester.
Castle, N.G., Mor, V., 1998. Physical restraint in nursing homes: a review of
the literature since the nursing home reform act of 1987. Medical Care
Research and Review 55 (2) 139–170.
Demir, A., 2007. Nurses’ use of physical restraint in four Turkish hospitals.
Journal of Nursing Scholarship 39 (1) 38–45.
Donner, A., Klar, N., 2000. Design and Analysis of Cluster Randomization
Trials in Health Research. Arnold, London.
Evans, D., Wood, J., Lambert, L., Fitzgerald, M., 2002. Physical restraint in
acute and residential care. A systematic review. No. 22, The Joanna
Briggs Institute, Adelaide, Australia Available from: http://www.joannabriggs.edu.au (accessed 22.03.12).
Feng, Z., Hirdes, J.P., Smith, T.F., Finne-Soveri, H., Ch, I., Du Pasquier, J.N.,
Gilgen, R., Ikegami, N., Mor, V., 2009. Use of physical restraints and
antipsychotic medications in nursing homes: a cross-national study.
International Journal of Geriatric Psychiatry 24 (10) 1110–1118.
Fogel, D., Steinert, T., 2012. Aggressive und gewalttätige Patienten –
Fixierung [Aggressive and violent patients – application of physical
restraints]. Lege Artis 2 (1) 28–33.
Goethals, S., Dierckx de Casterlé, B., Gastmans, C., 2012. Nurses’ decisionmaking in cases of physical restraint: a synthesis of qualitative
evidence. Journal of Advanced Nursing 68 (6) 1198–1210.
Hamers, J., Huizing, A., 2005. Why do we use physical restraints in the
elderly? Zeitschrift für Gerontologie und Geriatrie 38 (1) 19–25.
Healey, F.M., Cronberg, A., Oliver, D., 2009. Bedrail use in English and
Welsh hospitals. Journal of the American Geriatrics Society 57 (10)
1887–1891.
Heinze, C., Dassen, T., Grittner, U., 2012. Use of physical restraints in
nursing homes and hospitals and related factors: a cross-sectional
study. Journal of Clinical Nursing 21 (7–8) 1033–1040.
Janelli, L.M., Stamps, D., Delles, L., 2006. Physical restraint use: a nursing
perspective. Medsurg Nursing: Official Journal of the Academy of
Medical-Surgical Nurses 15 (3) 163–167.
Karlsson, S., Bucht, G., Sandman, P.O., 1998. Physical restraint in geriatric
care knowledge, attitudes and use. Scandinavian Journal of Caring
Sciences 12 (1) 48–56.
Köpke, S., Mühlhauser, I., Gerlach, A., Haut, A., Haastert, B., Möhler, R.,
Meyer, G., 2012. Effect of a guideline-based multicomponent intervention on use of physical restraints in nursing homes: a randomized
controlled trial. The Journal of the American Medical Association 307
(20) 2177–2184.
Krüger, C., Meyer, G., Hamers, J., 2010. Mechanische freiheitsentziehende
Maßnahmen im Krankenhaus: ein systematischer Literaturüberblick
G Model
8
[Physical restraints in hospital. A systematic overview] Zeitschrift für
Gerontologie und Geriatrie 43 (5) 291–296.
Lane, C., Harrington, A., 2011. The factors that influence nurses’ use of
physical restraints: a thematic literature review. International Journal
of Nursing Practice 17 (2) 195–204.
Laurin, D., Voyer, P., Verreault, R., Durand, P., 2004. Physical restraint use
among nursing home residents: a comparison of two data collection
methods. BMC Nursing 3, 5.
Le, Ch.T., 2003. Introductory Biostatistics. John Wiley & Sons, Inc., Hoboken, NJ.
Maccioli, G.A., Dorman, T., Brown, B.R., Mazuski, J.E., McLean, B.A., Kuszaj,
J.M., Rosenbaum, S.H., Frankel, L.R., Devlin, J.W., Govert, J.A., Smith, B.,
Peruzzi, W.T., American College of Critical Care Medicine, Society of
Critical Care Medicine, 2003. Clinical practice guidelines for the maintenance of patient physical safety in the intensive care unit: use of
restraining therapies. American College of Critical Care Medicine Task
Force 2001–2002. Critical Care Medicine 31 (11) 2665–2676.
Martin, B., 2002. Restraint use in acute and critical care settings: changing
practice. AACN Clinical Issues 13 (2) 294–306.
Martin, B., Mathisen, L., 2005. Use of physical restraints in adult care: a
bicultural study. American Journal of Critical Care 14 (2) 133–142.
Meyer, G., Köpke, S., Haastert, B., Mühlhauser, I., 2009. Restraint use
among nursing home residents: cross-sectional study and prospective cohort study. Journal of Clinical Nursing 18 (7) 981–990.
Meyers, H., Nikoletti, S., Hill, A., 2001. Nurses’ use of restraints and their
attitudes toward restraint use and the elderly in the acute care
setting. Nursing and Health Sciences 3 (1) 29–34.
Minnick, A.F., Mion, L.C., Johnson, M.E., Catrambone, C., Leipzig, R., 2007.
Prevalence and variation of physical restraint use in acute care
settings in the US. Journal of Nursing Scholarship 39 (1) 30–37.
Minnick, A.F., Mion, L.C., Leipzig, R., Lamb, K., Palmer, R.M., 1998. Prevalence and patterns of physical restraint use in the acute care setting.
Journal of Nursing Administration 28 (11) 19–24.
Park, M., Tang, J.H., 2007. Changing the practice of physical restraint use in
acute care. Journal of Gerontological Nursing 33 (2) 9–16.
Rechel, B., Wright, S., Edwards, N., Dowdeswell, B., McKee, M.
(Eds.), 2009. Investing in hospitals of the future. European
Observatory on Health Systems and Policies http://www.euro.
who.int/__data/assets/pdf_file/0009/98406/E92354.pdf.
Registered Nurses’ Association of Ontario (RNAO), 2012. Promoting
Safety: Alternative Approaches to the Use of Restraints. http://
rnao.ca/bpg/guidelines/promoting-safety-alternative-approachesuse-restraints.
Sokol, D.K., 2010. When is restraint appropriate? British Medical Journal
341, c4147.
Strout, T.D., 2010. Perspectives on the experience of being physically
restrained: an integrative review of the qualitative literature. International Journal of Mental Health Nursing 19 (6) 416–427.
Turgay, A.S., Sari, D., Genc, R.E., 2009. Physical restraint use in Turkish
intensive care units. Clinical Nurse Specialist 23 (2) 68–72.
Werner, P., Mendelsson, G., 2001. Nursing staff members’ intentions to
use physical restraints with older people: testing the theory of
reasoned action. Journal of Advanced Nursing 35 (5) 784–791.

Use of physical restraints in acute hospitals in Germany: A multi

Transcription

Similar documents

Es geht auch ohne freiheitsentziehende Maßnahmen