De Wvggz beoogt verplichte zorg zoveel mogelijk te voorkomen. Uit de eerste evaluatie van de Wvggz is echter gebleken dat het toepassen van verplichte zorg onvoldoende daalt. Momenteel heerst onduidelijkheid in de praktijk over de rol van preventiestrategieën en welke aanpak het meest effectief is met betrekking tot de preventie van verplichte zorg. Mede hierdoor is er sprake van praktijkvariatie. Deze module heeft als doel om effectieve strategieën te evalueren om verplichte zorg te voorkomen en/of te verminderen.

PICO 1 - Summary of Findings

Summary of Findings table: Prevention strategies compared to standard care or not applying a prevention strategy in patients with mental illness  

Population: Patients with mental illness

Intervention: Prevention strategy

Comparator: Standard care or not applying a prevention strategy

¹ Risk of Bias: serious. Due to missing information about randomization procedure, baseline differences between groups, missing outcome data, unclear risk of bias due to selective outcome reporting, unclear risk of bias due to lack of blinding. Indirectness: serious. Due to heterogeneity in study populations. Imprecision: serious. Due to wide confidence intervals of individual studies.

² Risk of Bias: serious. Due to missing information about randomization procedure, unclear risk of bias due to lack of blinding. Inconsistency: serious. Due to conflicting results. Indirectness: serious. Due to heterogeneity in study populations.

³ Risk of Bias: serious. Due to missing information about/insufficient randomization procedure, unclear risk of bias due to lack of blinding, missing information about loss to follow-up. Inconsistency: serious. Due to conflicting results. Indirectness: serious. Due to heterogeneity in study populations.

⁴ Risk of Bias: serious. Due to missing information about randomization procedure, missing outcome data, unclear risk of bias due to selective outcome reporting, unclear risk of bias due to lack of blinding. Indirectness: serious. Due to heterogeneity in study populations. Imprecision: serious. Due to wide confidence intervals of individual studies.

⁵ Risk of Bias: serious. Due to missing information about randomization procedure, unclear risk of bias due to lack of blinding. Indirectness: serious. Due to heterogeneity in study populations. Imprecision: serious. Due to wide confidence intervals of individual studies.

⁶ Risk of Bias: serious. Due to missing information about randomization procedure, missing outcome data, unclear risk of bias due to lack of blinding. Indirectness: serious. Due to heterogeneity in study populations. Imprecision: serious. Due to wide confidence intervals of individual studies.

⁷ Risk of Bias: serious. Due to missing information about/insufficient randomization procedure, unclear risk of bias due to lack of blinding, missing information about loss to follow-up. Imprecision: very serious. Due to missing dispersion measures and the inability to calculate effect measures.

*For further details, see risk of bias table in the appendix; ^· The number of participants included in Kontio (2014) is unclear.

PICO 2 - Summary of findings 

No studies were included and could therefore not be graded. 

Summary of literature

PICO 1 - Description of studies

A total of four systematic reviews and six RCTs were included in the analysis of the literature. Important characteristics of the included systematic reviews are summarized in table 2. Table 3 presents important study characteristics of the included RCTs. The assessment of the risk of bias is summarized in the risk of bias tables (under the tab ‘Evidence tabellen’).

Indregard (2024) performed a RCT to compare coercive practices in open-door policy and treatment as usual wards in an urban hospital setting. All patients (≥18 years) referred to the acute psychiatric ward at Lovisenberg Diaconal Hospital (Oslo, Norway) were screened for eligibility. Patients were excluded if 1) they were legally prohibited from admission in an open-door facility due to being under active criminal justice custody or serving a sentence, or 2) they had a documented history of persistent violence. Patients referred for acute psychiatric ward care from the admission ward were allocated to either an open-door policy ward (n=245) or a treatment as usual ward (n=311). In the open-door policy wards, the main doors were open from 9:00 to 21:00 every day of the week but could be locked if it was the only solution to ensure safety. Patients in the control wards received treatment as usual in which the main doors were always locked and patients’ freedom to leave the ward would depend on the staff’s availability and individual risk assessment. Staff on all wards underwent de-escalation and harm minimization training every three to six months. Additional study characteristics can be found in table 3.

Pedersen (2024) aimed to examine evaluated evidence-based interventions that seek to reduce the incidence of and/or time in mechanical restraint (restrictive practice in mental health to immobilize or restrict people’s movement in response to violent and aggressive behavior, defined as the use of equipment, such as belts, straps or cuffs, by trained staff) in adult mental health inpatient settings. To identify relevant studies, a two-phased approach was used. First, the review by Baker (2021) was reviewed to identify relevant studies between 1999 and April 2019. Second, an updated systematic search was performed from 1 January 2019 to 4 July 2023 based on the search strategy adopted in Baker (2021). The international databases CINAHL, Embase, and PubMed were searched. Inclusion criteria were: 1) peer-reviewed primary research literature, 2) evaluation of interventions that seek to reduce the incidence of and/or time in mechanical restraint (or seclusion/mechanical restraint reported simultaneously), and 3) English full-text available. Studies that also included findings for children or adolescents or other types of restraint, but which were not reported independently, were excluded. In total, 41 studies were included in the review. Of these, five studies were controlled trials (Celofiega, 2022; Hamann, 2020; Hvidhjelm, 2022; Kontio, 2014; Putkonen, 2013) and were included in the current analysis.

Steinert (2023) performed a RCT to investigate whether the number of coercive measures used in psychiatric wards could be reduced by implementation of recommendations of the PreVCo (prevention of violence and coercion) program. Wards were eligible to participate in the study if they: 1) admitted involuntary patients regardless of their diagnoses, 2) routinely recorded data on coercive measures and incidents of aggression, and 3) had the support of both medical and nursing management for the study's objectives. A total of 54 wards were considered eligible and randomized to either an intervention (n=27 wards) or a control condition (n=27 wards) (waiting list). The intervention consisted of the implementation of three out of 12 suggested recommendations or at least one complex intervention program drawn from the guidelines (tailored approach). Workshops were conducted with the ward teams for implementation. Additional study characteristics can be found in table 3.

Cornelis (2022) conducted a RCT to evaluate the potential of intensive home treatment to reduce bed-days and its clinical effectiveness compared with treatment as usual. Participants were recruited at the psychiatric emergency service in Amsterdam and from emergency wards of two major mental health institutions Arkin and GGZ inGeest. Inclusion criteria were: 1) a diagnosis of at least one DSM-IV-TR or DSM-5 disorder (not primarily a substance use disorder), 2) age 18-65 years, and 3) experiencing an acute severe psychiatric crisis for which a psychiatrist indicated or executed a clinical admission. Participants that were receiving Flexible Assertive Community Treatment (FACT) with integrated crisis care or participants that received previous intensive home treatment were excluded. Eligible participants were randomly assigned to either the intensive home treatment group (n=183), or the treatment as usual group (n=63). Intensive home treatment provided intensive outpatient care and functioned as a gatekeeper for hospital admissions. Treatment plans were tailored to the patient and included for example pharmacotherapy, structuring the daytime, psychoeducation, cognitive behavioral therapy, and social care. Participants in the treatment as usual group received a combination of highly intensive inpatient treatment in the first phase, and low frequency outpatient treatment in the second phase. Treatment settings included hospital wards, acute treatment teams, and outpatient clinics (no home treatment). Additional study characteristics are presented in table 3.

Tinland (2022) conducted a RCT to determine the efficacy of psychiatric advance directives (PADs) facilitated by peer workers in people with mental disorders. Participants were included if they 1) were aged older than 18 years, 2) had a diagnosis of schizophrenia, bipolar disorder, or schizoaffective disorder according to the DSM-5, 3) were involuntary admitted to the hospital within the past 12 months, and 4) had decision-making capacity according to the MacArthur Competence Assessment Tool for Clinical Research. Eligible participants were randomized to either the peer-worker psychiatric advance directives (PW-PAD) group (n=196), or the control group (n=198). Participants in the PW-PAD group could, depending on their preferences, meet a peer worker in a place of their choice, be supported by this peer worker in drafting the PAD document, and/or be supported by the peer worker during the sharing of PADs with the health care agent and the psychiatrist. Participants assigned to the control group were followed up as usual. Additional study characteristics can be found in table 3.

Välimäki (2022) performed a cluster-RCT to examine whether an evidence-based educational intervention for nurses decreases the use of seclusion rooms in psychiatric hospitals compared with usual practice. Hospitals with at least one adult psychiatric ward that used coercive measures as defined in the Finnish Mental Health Act were eligible. Hospital wards that specialized solely in child and adolescent psychiatric care, psychogeriatric care, or forensic psychiatry were excluded. All nurses working on the study wards were invited to join the study, without any in- or exclusion criteria. For patients, inclusion criteria were: 1) admission to the study ward, 2) aged ≥18 years, and 3) fluency in Finnish. Hospitals were randomly assigned to the evidence-based educational intervention (VIOLIN) group (n=13 wards) or the usual practice group (n=15 wards). VIOLIN is an 18-month intervention aimed to reduce coercive practices in psychiatric wards. Eight months of intervention were followed by a 10-month maintenance period. Usual practice wards continued with their normal activities. Additional study characteristics are displayed in table 3.

Baker (2021) conducted a systematic review to provide an overview of non-pharmacological interventions to reduce restrictive practices in adult mental health inpatient settings. The following databases were searched from 1999 to April 2019: British Nursing Index (BNI), CINAHL, Cochrane Central Register of Controlled Trials (CENTRAL), Cochrane Database of Systematic Reviews (CDSR), Database of Abstracts of Reviews of Effects (DARE), Embase, Health Technology Assessment (HTA) database, HTA Canadian and International, Medline, NHS Economic Evaluation Database (NHS EED), PsycINFO, and PubMed. Additionally, a grey literature search was performed. Inclusion criteria were: 1) evaluation of interventions aimed at changing the behavior of inpatient adult mental health service staff to reduce restrictive practices, 2) reporting on the outcome reduction of restrictive practices, and 3) English full-text available. Studies that included findings regarding children and adolescent mental health services, learning disability and organic conditions or studies that evaluated pharmacological interventions were excluded. In total, 175 studies were included in the review of which six studies used a RCT design. Eventually, two RCTs were included in the current analysis (van de Sande, 2011; Abderhalden, 2008). Two RCTs were excluded because they were included in Pedersen (2024) (Kontio, 2014; Putkonen, 2013), one because no relevant outcomes -according to the PICO- were reported (Bowers, 2015), and one because no full-text was available (Parasurum, 2011).

Hamann (2020) performed a post-hoc analysis with data from the cluster-randomized SDM-plus trial to investigate whether shared decision making (SDM) reduces the number of patients with incidents of aggression and the use of coercive measures compared to treatment as usual. The study was conducted in 12 acute psychiatric wards of 4 psychiatric hospitals in Germany. Inclusion criteria were: 1) inpatient status of participating ward, 2) age 18-65 years, and 3) diagnosis of schizophrenia or schizoaffective disorder. Patients with intellectual disabilities were excluded. Pairs of comparable wards were determined, after which one ward of each pair was randomized to the intervention group (n=154) and the other to the control group (n=151). In the intervention group, staff, and patients both received training in SDM-plus, which aims to empower health care staff and patients regarding SDM-specific communication techniques. Staff and patients from the control wards acted under treatment as usual conditions. They were offered SDM-plus training after the end of the study. Additional study characteristics are presented in table 3.

Molyneaux (2019) aimed to synthesize evidence on the effectiveness of crisis-planning interventions in reducing rates of compulsory hospital admissions for people with psychotic illness or bipolar disorder, compared with usual care. The international databases CENTRAL, CINAHL, Medline, Embase, PsycINFO, and the international Standard Randomized Controlled Trial Number (ISRCTN) registry were searched from inception to 16 October 2018. Inclusion criteria were: 1) RCTs, 2) adults (≥16 years) with a diagnosis of a psychotic illness or bipolar disorder with or without psychotic symptoms, 3) examination any form of crisis-planning intervention, and 3) comparison group was treatment as usual. There were no exclusion criteria based on language or date of publication. A total of five RCTs were included in the review, which were all included in the current analysis (Papageorgiou, 2002; Henderson, 2004; Thornicroft, 2013; Ruchlewska, 2014; Lay, 2018).

De Jong (2016) performed a systematic review to establish which interventions effectively reduce compulsory admissions in adult psychiatric patients in outpatient settings. The international databases Embase, Medline, Web of Science, PsycINFO, CINAHL, and CENTRAL were searched from inception to 30 April 2015. Studies were eligible if: 1) they were RCTs, 2) they investigated interventions of any kind that were designed to reduce compulsory admission rates in adult psychiatric patients in outpatient settings, 3) they reported on compulsory admission as primary or secondary outcome measure. Studies with compulsory admission as an intervention or any intervention during hospitalization were not eligible. A total of thirteen RCTs were included in the review, of which eight were included in the current analysis (Swartz, 1999; Steadman, 2001; Burns, 2013; Staring, 2010; Priebe, 2013; Johnson, 2005; Ohlenschlaeger, 2008; Lay, 2015). Four RCTs were excluded because they were included in Molyneaux (2019) (Papageorgiou, 2002; Henderson, 2004; Thornicroft, 2013; Ruchlewska, 2014), and one because of its long follow-up of 12 years (Sigrunarson, 2013).

PICO 1 - Results

1. Compulsory admission

Fifteen studies reported on the outcome measure compulsory admissions, defined as the number of compulsorily admitted patients (Burns, 2013; Henderson, 2004; Johnson, 2005; Lay, 2015; Lay, 2018; Ohlenschlaeger, 2008; Papageorgiou, 2002; Priebe, 2013; Ruchlewska, 2014; Staring, 2010; Steadman, 2001; Swartz, 1999; Thornicroft, 2013; Cornelis, 2022; Tinland, 2022).

The results were pooled using four subgroups based on the kind of prevention strategy that was studied (advance statements, community treatment orders, compliance enhancement, and intensive home treatment) (figure 1). Results of Johnson (2005), Lay (2015), and Ohlenschlaeger (2008) could not be pooled due to heterogeneity in the interventions studied.

The pooled data of RCTs studying advance statements show a risk ratio of 0.73 (95%CI 0.61 to 0.86) in favor of advance statements. The risk of compulsory admission had been reduced by 27%. This difference was considered clinically relevant. The pooled data of RCTs studying community treatment orders show a risk ratio of 0.95 (95%CI 0.81 to 1.10) favoring community treatment orders. This difference was considered not clinically relevant. The pooled data of RCTs studying compliance enhancement show a risk ratio of 0.52 (95%CI 0.11 to 2.37) in favor of compliance enhancement. The risk of compulsory admission had been reduced by 48%, which was considered a clinically relevant difference.

Figure 1: The effect of prevention strategies compared with standard care on compulsory admissions.

Z: p-value of the pooled effect; df: degrees of freedom; I2: statistic heterogeneity; CI: confidence interval.

Cornelis (2022) studied intensive home treatment reported that 62.3% (114/183) of participants in the intervention group (intensive home treatment) was compulsory admitted, compared to 73.0% (46/63) of participants in the control group (standard care). Risk ratio was 0.85 (95%CI 0.71 to 1.03), in favor of intensive home treatment. This difference is considered clinically relevant.

Johnson (2005) reported that 17.8% (24/135) of participants in the intervention group (crisis resolution) was compulsorily admitted, compared to 25.6% (32/125) of participants in the control group (standard care). Risk ratio was 0.69 (95%CI 0.43 to 1.11) in favor of crisis resolution. This difference is considered clinically relevant.

Lay (2015) found that 18 out of 80 (22.5%) participants in the intervention group (psychoeducation) were compulsorily admitted, compared to 36 out of 102 (35.3%) participants in the control group (treatment as usual). Risk ratio was 0.64 (95%CI 0.39 to 1.03) in favor of psychoeducation. This difference is considered clinically relevant.

Ohlenschlaeger (2008) reported that 16.8% (28/167) of participants in the intervention group (integrated care) was compulsorily admitted, compared to 14.3% (23/161) of participants in the control group (standard care). Risk ratio was 1.17 (95%CI 0.71 to 1.95) in favor of standard care. This difference is considered clinically relevant.

2. Physical and mechanical restraint

Seven studies reported on the incidence of physical and mechanical restraint (Celofiega, 2022; Hvidhjelm, 2022; Kontio, 2014; Ohlenschlaeger, 2008; Indregard, 2024; Valimaki, 2022; Hamann, 2020). Results could not be pooled due to heterogeneity in interventions and reporting of the outcome measure. For one study (Kontio, 2014), effect sizes ([standardized] mean difference/risk ratio) could not be calculated, because mean change values were reported.

Celofiega (2022) reported on the incidence of mechanical restraint (with belts) during the intervention period. The authors reported the use of mechanical restraint in 7.4% (100/1347) of patients in the de-escalation education group, and in 11.4% (213/1864) of patients in the treatment as usual group. Risk ratio was 0.65 (95%CI 0.52 to 0.82) in favor of de-escalation education (figure 2). This difference is considered clinically relevant.

Hvidhjelm (2022) reported on the mean incidence rate (number/month) of mechanical restraint. The mean (SD) rate of mechanical restraint use within the intervention period was 0.03 (0.10) for the START group (n=42) and 0.18 (0.39) for the control group (n=94). Mean difference was 0.15 (95%CI -0.23 to -0.07) in favor of START. This difference is considered clinically relevant.

Kontio (2014) reported mean (SD) changes in rates (incidents per 1000 occupied bed days) of mechanical restraint. The authors reported a mean change in the rates of mechanical restraint incidents of 2.9 (3.7) for the e-learning course wards, and 0.9 (1.0) for the treatment as usual wards.

Ohlenschlaeger (2008) reported on the number of patients mechanically restrained (use of a leather belt) and the number of patients physically restrained (use of force). The authors reported mechanical restraint in 7% (11/167) of patients in the integrated care group, and in 5% (8/161) of patients in the standard care group. Risk ratio was 1.33 (95%CI 0.55 to 3.21) in favor of standard care (figure 2). This difference is considered clinically relevant. Additionally, the authors reported physical restraint in 5% (9/167) of patients in the integrated care group, and in 3% (5/161) of patients in the standard care group. Risk ratio was 1.74 (95%CI 0.59 to 5.07) favoring standard care (figure 2). This difference is also considered clinically relevant.

Indregard (2024) reported on the number of patients mechanically restrained. The authors reported mechanical restraint in 3.3% (8/245) of patients in the open-door policy wards, and in 4.8% (15/311) of patients in the treatment as usual wards. Risk ratio was 0.68 (95%CI 0.29 to 1.57) favoring open-door policy (figure 2). This difference is considered clinically relevant. In addition, the authors also reported on physical restraint: 10.2% (25/245) in the open-door policy wards, and 12.9% (40/311) in the treatment as usual wards. Risk ratio was 0.79 (95%CI 0.50 to 1.27) in favor of open-door policy (figure 2). This difference is considered clinically relevant.

Valimaki (2022) reported on the number of patients physically restrained. Use of physical restraint was reported for 70 out of 4089 (1.7%) patients in the intervention group (VIOLIN) compared to 14 out of 4092 (0.3%) patients in the treatment as usual group. Risk ratio was 5.00 (95%CI 2.82 to 8.87) favoring treatment as usual, which was considered clinically relevant.

Hamann (2020) reported on the number of patients being mechanically restrained at least once during the intervention phase. Mechanical restraint was reported in 5% (7/154) of patients in the SDM-plus training group, and in 3% (4/151) of patients in the treatment as usual group. Risk ratio was 1.72 (95%CI 0.51 to 5.74) in favor of treatment as usual (figure 2). This difference was considered clinically relevant.

Figure 2: The effect of prevention strategies compared with standard care on physical and mechanical restraint. 

Z: p-value of the pooled effect; df: degrees of freedom; I2: statistic heterogeneity; CI: confidence interval.

3. Seclusion

Five studies reported on the outcome measure seclusion (Kontio, 2014; Ohlenschlaeger, 2008; van de Sande, 2011; Indregard, 2024; Valimaki, 2022). Results could not be pooled due to heterogeneity in interventions and reporting of the outcome measures. Effect sizes ([standardized] mean difference/risk ratio) could not be calculated for Kontio (2014), because mean change values were reported.

Kontio (2014) reported mean (SD) changes in rates (incidents per 1000 occupied bed days) of seclusion. The authors reported a mean change in the rates of seclusion of 15.6 (9.2) for the e-learning course wards, and 8.7 (4.4) for the treatment as usual wards.

Ohlenschlaeger (2008) reported on the number of patients that were secluded (locked door). The authors reported that none of the patients in the integrated treatment group (n=167) were secluded, compared to three patients (2%) in the standard treatment group (n=161). Risk ratio was 0.14 (95%CI 0.01 to 2.65) favoring integrated treatment. This difference was considered clinically relevant.

Van de Sande (2011) reported on the number of patients that were secluded. The authors reported that 29% (60/207) of patients in the intervention wards (structured short-term risk assessment) were secluded compared to 17% (42/251) of patients in the control wards (treatment as usual). Risk ratio was 1.73 (95%CI 1.22 to 2.46) favoring treatment as usual. This difference was considered clinically relevant.

Indregard (2024) reported on the number of secluded patients. 5.3% (13/245) of patients in the open-door policy wards were secluded compared to 5.5% (17/311) of patients in the treatment as usual wards. Risk ratio was 0.97 (95%CI 0.48 to 1.96) favoring open-door policy. This difference was considered not clinically relevant.

Valimaki (2022) reported on the number of patients that were secluded. The authors reported that 437 out of 4089 (10.7%) of patients in the educational intervention group were secluded, and 512 out of 4092 (12.5%) of patients in the treatment as usual group. Risk ratio was 0.85 (95%CI 0.76 to 0.96), in favor of the educational intervention. This difference was considered clinically relevant.

4. Duration of compulsory admission

Six studies reported on the outcome measure duration of compulsory admission (Burns, 2013; Johnson, 2005; Lay, 2015; Lay, 2018; Thornicroft, 2013; Cornelis, 2022). Thornicroft (2013), Lay (2015) and Lay (2018) reported on the mean duration of compulsory admissions in days. Burns (2013) reported on the mean total duration of all psychiatric hospitalizations in nights. Johnson (2005) and Cornelis (2022) reported on mean bed days in hospitals/crisis house.

Results of Lay (2018) and Thornicroft (2013) were pooled (figure 3). Results of Johnson (2005) and Lay (2015) could not be pooled due to heterogeneity in the interventions studied. Results of Burns (2013) and Cornelis (2022) could not be pooled due to results of one study only.

The pooled data of RCTs studying advance statements show a standardized mean difference (SMD) of -0.13 (95%CI -0.48 to 0.22) in favor of advance statements. This difference was considered not clinically relevant.

Figure 3: The effect of prevention strategies compared with standard care on the duration of compulsory admission.

Z: p-value of the pooled effect; df: degrees of freedom; I2: statistic heterogeneity; CI: confidence interval.

Johnson (2005) reported 21.3 (37.9) mean (SD) bed days in a hospital/crisis house for patients the intervention group (crisis resolution) compared to 38.6 (47) days for patients in the control group (standard care). Mean difference was 17.30 (95%CI -27.73 to -6.87) in favor of crisis resolution. This difference is considered clinically relevant.

Lay (2015) reported a mean (SD) duration of compulsory admission of 9.1 (21.8) days for the intervention group (psychoeducation) compared to 14.8 (31.2) days for the control group (treatment as usual). Mean difference was 5.70 (95%CI -13.41 to 2.01) in favor of psychoeducation. This difference is considered clinically relevant.

Burns (2013) reported a mean (SD) total duration of all psychiatric hospitalizations of 82.2 (102) nights for the intervention group (community treatment order) compared to 90.9 (104.5) nights for the control group (standard outpatient care). Mean difference was 8.70 (95%CI -30.88 to 13.48) in favor of community treatment order. This difference is considered not clinically relevant.

Cornelis (2022) reported on mean (SD) bed days in hospital. The authors reported 18.1 (17.7) mean bed days in hospital for the intervention group (intensive home treatment) and 24.2 (17.5) mean bed days in hospital for the control group (treatment as usual). Mean difference was 6.10 (95%CI -11.12 to -1.08) in favor of intensive home treatment. This difference is considered clinically relevant.

5. Duration of physical and mechanical restraint

Three studies reported on the duration of mechanical restraint (Hvidhjelm, 2022; Kontio, 2014; Ohlenschlaeger, 2008). Results could not be pooled due to variability in reporting of the outcome measure. For one study (Kontio, 2014), effect sizes ([standardized] mean difference/risk ratio) could not be calculated, because mean change values were reported.

Hvidhjelm (2022) reported on the mean (SD) duration of mechanical restraint. A mean mechanical restraint duration of 274 (1652) minutes was reported for the START group (n=42) compared to 609 (1950) minutes for the control group (n=94). Mean difference was 335 (95%CI -971.40 to 301.40) favoring START. This difference was considered not clinically relevant.

Kontio (2014) reported mean (SD) changes in time patient spent in mechanical restraint. The authors reported a mean change in time patient spent in mechanical restraint of 27.4 (67.2) for the e-learning course wards, and 26.7 (34.9) for the treatment as usual wards.

Ohlenschlaeger (2008) reported on the mean (SD) duration of mechanical restraint with the leather belt. They reported a mean duration of 19.4 (24.2) hours for the integrated care group (n=167), and 37.4 (50.3) hours in the standard care group (n=161). Mean difference was 18.0 (95%CI -26.59 to -9.41) in favor of integrated care. This difference was considered clinically relevant.

6. Compulsory care

Six studies reported on the outcome measure compulsory care, using different definitions (Abderhalden, 2008; Hvidhjelm, 2022; Ohlenschlaeger, 2008; Putkonen, 2013; Indregard, 2024; Steinert, 2023). Results could not be pooled due to variability in reporting of the outcome measure. For three studies (Abderhalden, 2008; Putkonen, 2013; Steinert, 2023), effect sizes ([standardized] mean difference/risk ratio) could not be calculated, because incidence rate(s) (ratio’s) or median (IQR) values were reported.

Abderhalden (2008) reported on compulsory care, defined as the overall incidence rate of coercion during the intervention period. The overall incidence rate of coercion was 1.75 (95%CI 1.47 to 2.07) per 100 hospitalization days in the intervention group (structured short-term risk assessment) (135 incidents involving 390 patients), and 1.20 (95%CI 1.00 to 1.43) per 100 hospitalization days in the control group (126 incidents involving 583 patients).

Bowers (2015) did not report sufficient results for GRADE evaluation; these results were therefore not graded.

Hvidhjelm (2022) reported on total coercive episodes, defined as the number of physical restraint episodes, episodes of acute forced medication, episodes of one-to-one observation, and mechanical restraint episodes. The mean (SD) rate of total use of coercion was 0.12 (0.32) for the START group (n=42), and 0.39 (0.94) for the control group (n=94). Mean difference was 0.27 (95%CI -0.48 to -0.06) in favor of START. This difference was considered clinically relevant.

Putkonen (2013) reported on compulsory care, defined as rates of seclusion, restraints, and room observation. The authors reported an incidence rate ratio of 0.88 (95%CI 0.86 to 0.90) for six core strategies wards, and 0.97 (95%CI 0.93 to 1.01) for control wards.

Ohlenschlaeger (2008) reported on the total number of patients subjected to coercive measure(s). 19% (31/167) of patients in the integrated care group were subjected to coercive measure(s) compared to 16% (25/161) of patients in the standard care group. Risk ratio was 1.20 (95%CI 0.74 to 1.93) favoring standard care. This difference was considered clinically relevant.

Indregard (2024) reported on the total number of patients subjected to one or more coercive measures during admission. The authors reported that 65 out of 245 (26.5%) of patients in the open-door policy wards were subjected to one or more coercive measures compared to 104 out of 311 (33.4%) of patients in the treatment as usual wards. Risk ratio was 0.79 (95%CI 0.61 to 1.03) in favor of open-door policy. This difference was considered clinically relevant.

Steinert (2023) reported on the median (IQR) number of coercive measures used per month and occupied bed. The authors reported a value of 0.53 (0.59) for the intervention wards (n=27) and 0.71 (1.08) for the waiting list wards (n=27).

7. Duration of seclusion

Three studies reported on duration of seclusion (Kontio, 2014; van de Sande, 2011; Valimaki, 2022). Effect sizes ([standardized] mean difference/risk ratio) could not be calculated, because mean change values were reported (Kontio, 2014) or due to missing dispersion measures (SD/SE) (van de Sande, 2011; Valimaki, 2022).

Kontio (2014) reported on mean (SD) changes in time patient spent in seclusion. A mean change of 42.8 (86.9) was reported for patients in the e-learning course wards, and 27.4 (67.2) for patients in the treatment as usual wards.

Van de Sande (2011) reported on seclusion duration. The authors reported a seclusion duration of 1624 hours during the intervention period for the structured short-term risk assessment wards and 2149 hours for the treatment as usual wards.

Valimaki (2022) reported on the length per seclusion event on ward level, expressed as geometric mean. The authors reported a seclusion length of 1762 minutes for the educational intervention group and 1614 minutes for the treatment as usual group.

8. Isolation (prevention, duration), 9. Care authorization

None of the studies reported on the outcome measures isolation and care authorization.

PICO 2 - Results 

No studies were included in the literature analysis

Search and select

A systematic review of the literature was performed to answer the following question(s):

1. What is the effectiveness of a prevention strategy compared to not applying a prevention strategy or standard care (‘treatment as usual’) in preventing compulsory care for patients with mental illness?
2. What is the most effective prevention strategy to prevent compulsory care for patients with mental illness?

Table 1. PICO

Relevant outcome measures

The guideline panel considered compulsory admission, physical and mechanical restraint, and seclusion as critical outcome measures for decision making; and duration of compulsory admission, duration of physical and mechanical restraint, compulsory care, duration of seclusion, isolation, and care authorization as important outcome measures for decision making.

A priori, the guideline panel did not define the outcome measures listed above but used the definitions used in the studies.

For all outcome measures the guideline panel defined a 10% difference for dichotomous outcomes (0.9 ≥ RR ≥ 1.1) and 0.2 SD or -0.2 ≥ SMD ≥ 0.2 for continuous outcomes as a minimal clinically (patient) important difference.

Search and select (Methods)

The databases Medline (via OVID), Embase (via Embase.com), and PsycINFO (via OVID) were searched with relevant search terms from 2005 until 9 September 2024. The detailed search strategy is listed under the tab ‘Literature search strategy.’ The systematic literature search resulted in 405 hits. Studies were selected based on the following criteria:

• Systematic reviews in which searches were performed in at least two databases, with a detailed search strategy, risk of bias assessment and results of individual studies available;
• Full-text English or Dutch language publication; and
• Studies according to the PICO.

Initially, 29 systematic reviews were selected based on title and abstract screening. After reading the full text, 25 systematic reviews were excluded (see the exclusion table under the tab ‘Evidence tabellen’), and four were selected for inclusion. Additionally, RCTs with a publication date of 2018 or more recent (based upon the last search date of the systematic review performed by Molyneaux, 2019) were screened. Twenty RCTs were selected based on title and abstract screening. After reading the full text, fourteen RCTs were excluded (see the exclusion table), and six were selected for inclusion.

All included studies compared a prevention strategy to standard care or not applying a prevention strategy (comparison 1), no studies were included that compared two different prevention strategies (comparison 2). Thus, the literature summary focuses specifically on comparison 1.

1. Abderhalden C, Needham I, Dassen T, Halfens R, Haug HJ, Fischer JE. Structured risk assessment and violence in acute psychiatric wards: randomised controlled trial. Br J Psychiatry. 2008 Jul;193(1):44-50. doi: 10.1192/bjp.bp.107.045534. PMID: 18700217.
2. Ali R, Walker S, Nyikavaranda P, Downs J, Patel R, Khondoker M, Bhui K, Hayes RD, de Freitas DF. Ethnicity and involuntary hospitalisation: a study of intersectional effects. Soc Psychiatry Psychiatr Epidemiol. 2025 Sep;60(9):2061-2075. doi: 10.1007/s00127-025-02898-0. Epub 2025 Apr 16. PMID: 40240693; PMCID: PMC12378639.
3. Aragonés-Calleja M, Sánchez-Martínez V. Evidence synthesis on coercion in mental health: An umbrella review. Int J Ment Health Nurs. 2024 Apr;33(2):259-280. doi: 10.1111/inm.13248. Epub 2023 Oct 31. PMID: 37908175.
4. Barnett P, Mackay E, Matthews H, Gate R, Greenwood H, Ariyo K, Bhui K, Halvorsrud K, Pilling S, Smith S. Ethnic variations in compulsory detention under the Mental Health Act: a systematic review and meta-analysis of international data. Lancet Psychiatry. 2019 Apr;6(4):305-317. doi: 10.1016/S2215-0366(19)30027-6. Epub 2019 Mar 4. PMID: 30846354; PMCID: PMC6494977.
5. Barrett B, Waheed W, Farrelly S, Birchwood M, Dunn G, Flach C, Henderson C, Leese M, Lester H, Marshall M, Rose D, Sutherby K, Szmukler G, Thornicroft G, Byford S. Randomised controlled trial of joint crisis plans to reduce compulsory treatment for people with psychosis: economic outcomes. PLoS One. 2013 Nov 25;8(11):e74210. doi: 10.1371/journal.pone.0074210. PMID: 24282495; PMCID: PMC3839936.
6. Burns T, Rugkåsa J, Molodynski A, Dawson J, Yeeles K, Vazquez-Montes M, Voysey M, Sinclair J, Priebe S. Community treatment orders for patients with psychosis (OCTET): a randomised controlled trial. Lancet. 2013 May 11;381(9878):1627-33. doi: 10.1016/S0140-6736(13)60107-5. Epub 2013 Mar 26. PMID: 23537605.
7. Celofiga A, Kores Plesnicar B, Koprivsek J, Moskon M, Benkovic D, Gregoric Kumperscak H. Effectiveness of De-Escalation in Reducing Aggression and Coercion in Acute Psychiatric Units. A Cluster Randomized Study. Front Psychiatry. 2022 Apr 7;13:856153. doi: 10.3389/fpsyt.2022.856153. PMID: 35463507; PMCID: PMC9021532.
8. Cornelis J, Barakat A, Blankers M, Peen J, Lommerse N, Eikelenboom M, Zoeteman J, Van H, Beekman ATF, Dekker J. The effectiveness of intensive home treatment as a substitute for hospital admission in acute psychiatric crisis resolution in the Netherlands: a two-centre Zelen double-consent randomised controlled trial. Lancet Psychiatry. 2022 Aug;9(8):625-635. doi: 10.1016/S2215-0366(22)00187-0. Epub 2022 Jun 30. PMID: 35779532.
9. Daguman EI, Hutchinson M, Lakeman R. Uncovering Complexities in Reducing Aggression, Conflict and Restrictive Practices in Acute Mental Healthcare Settings: An Overview of Reviews. Int J Ment Health Nurs. 2024 Dec;33(6):1666-1686. doi: 10.1111/inm.13376. Epub 2024 Jun 17. PMID: 38886873.
10. Doedens P, Vermeulen J, Boyette LL, Latour C, de Haan L. Influence of nursing staff attitudes and characteristics on the use of coercive measures in acute mental health services-A systematic review. J Psychiatr Ment Health Nurs. 2020 Aug;27(4):446-459. doi: 10.1111/jpm.12586. Epub 2020 Jan 14. PMID: 31876970; PMCID: PMC7508163.
11. Efkemann SA, Scholten M, Bottlender R, Juckel G, Gather J. Influence of mental health professionals' attitudes and personality traits on decision-making around coercion: Results from an experimental quantitative survey using case vignettes. Acta Psychiatr Scand. 2022 Aug;146(2):151-164. doi: 10.1111/acps.13429. Epub 2022 Apr 4. PMID: 35322402.
12. Gooding P. Compendium report - Good practices to promote Voluntary Measures in Mental Health Services. 2021. Council of Europe. Beschikbaar via: https://rm.coe.int/inf-2021-9-compendium-final-e/1680b11f60, geraadpleegd op 9 september 2025
13. Hamann J, John M, Holzhüter F, Siafis S, Brieger P, Heres S. Shared decision making, aggression, and coercion in inpatients with schizophrenia. Eur Psychiatry. 2020 Sep 28;63(1):e90. doi: 10.1192/j.eurpsy.2020.88. PMID: 32981554; PMCID: PMC7576528.
14. Henderson C, Flood C, Leese M, Thornicroft G, Sutherby K, Szmukler G. Effect of joint crisis plans on use of compulsory treatment in psychiatry: single blind randomised controlled trial. BMJ. 2004 Jul 17;329(7458):136. doi: 10.1136/bmj.38155.585046.63. Epub 2004 Jul 7. PMID: 15240438; PMCID: PMC478218.
15. Hvidhjelm J, Brandt-Christensen M, Delcomyn C, Møllerhøj J, Siersma V, Bak J. Effects of Implementing the Short-Term Assessment of Risk and Treatability for Mechanical Restraint in a Forensic Male Population: A Stepped-Wedge, Cluster-Randomized Design. Front Psychiatry. 2022 Feb 24;13:822295. doi: 10.3389/fpsyt.2022.822295. PMID: 35280154; PMCID: PMC8907583.
16. Indregard AR, Nussle HM, Hagen M, Vandvik PO, Tesli M, Gather J, Kunøe N. Open-door policy versus treatment-as-usual in urban psychiatric inpatient wards: a pragmatic, randomised controlled, non-inferiority trial in Norway. Lancet Psychiatry. 2024 May;11(5):330-338. doi: 10.1016/S2215-0366(24)00039-7. Epub 2024 Mar 6. PMID: 38460529.
17. Inspectie voor Gezondheidszorg en Jeugd. Toetsingskader terugdringen insluiting. 2025. Beschikbaar via: https://www.igj.nl/documenten/2025/03/05/toetsingskader-terugdringen-insluiting, geraadpleegd op 14 november 2025
18. Johnson S, Nolan F, Pilling S, Sandor A, Hoult J, McKenzie N, White IR, Thompson M, Bebbington P. Randomised controlled trial of acute mental health care by a crisis resolution team: the north Islington crisis study. BMJ. 2005 Sep 17;331(7517):599. doi: 10.1136/bmj.38519.678148.8F. Epub 2005 Aug 15. PMID: 16103032; PMCID: PMC1215550.
19. Kinane C, Osborne J, Ishaq Y, Colman M, MacInnes D. Peer supported Open Dialogue in the National Health Service: implementing and evaluating a new approach to Mental Health Care. BMC Psychiatry. 2022 Feb 22;22(1):138. doi: 10.1186/s12888-022-03731-7. PMID: 35193551; PMCID: PMC8862567.
20. Kontio R, Pitkänen A, Joffe G, Katajisto J, Välimäki M. eLearning course may shorten the duration of mechanical restraint among psychiatric inpatients: a cluster-randomized trial. Nord J Psychiatry. 2014 Oct;68(7):443-9. doi: 10.3109/08039488.2013.855254. Epub 2013 Nov 25. PMID: 24274836.
21. Lay B, Blank C, Lengler S, Drack T, Bleiker M, Rössler W. Preventing compulsory admission to psychiatric inpatient care using psycho-education and monitoring: feasibility and outcomes after 12 months. Eur Arch Psychiatry Clin Neurosci. 2015 Apr;265(3):209-17. doi: 10.1007/s00406-014-0553-1. Epub 2014 Nov 1. PMID: 25361537.
22. Lay B, Kawohl W, Rössler W. Outcomes of a psycho-education and monitoring programme to prevent compulsory admission to psychiatric inpatient care: a randomised controlled trial. Psychol Med. 2018 Apr;48(5):849-860. doi: 10.1017/S0033291717002239. Epub 2017 Aug 14. PMID: 28805175.
23. Lim E, Wynaden D, Heslop K. Recovery-focussed care: How it can be utilized to reduce aggression in the acute mental health setting. Int J Ment Health Nurs. 2017 Oct;26(5):445-460. doi: 10.1111/inm.12378. PMID: 28960737.
24. Molenaar H, Roode R.P. de, Attema S.J. Eigen regie en de Wvggz. JGGZR+. 2025 Sept;3:20-24.
25. Moss K, Wyder M, Braddock V, Arroyo D, Kisely S. Compulsory community treatment and ethnicity: Findings from a culturally and linguistically diverse area of Queensland. Int J Law Psychiatry. 2019 Jan-Feb;62:154-159. doi: 10.1016/j.ijlp.2018.09.007. Epub 2018 Sep 28. PMID: 30274852.
26. Ohlenschlaeger J, Nordentoft M, Thorup A, Jeppesen P, Petersen L, Christensen TØ, Krarup G, Jørgensen P. Effect of integrated treatment on the use of coercive measures in first-episode schizophrenia-spectrum disorder. A randomized clinical trial. Int J Law Psychiatry. 2008 Jan-Feb;31(1):72-6. doi: 10.1016/j.ijlp.2007.11.003. Epub 2008 Jan 9. PMID: 18191455.
27. Papageorgiou A, King M, Janmohamed A, Davidson O, Dawson J. Advance directives for patients compulsorily admitted to hospital with serious mental illness. Randomised controlled trial. Br J Psychiatry. 2002 Dec;181:513-9. doi: 10.1192/bjp.181.6.513. PMID: 12456522.
28. Plomp E. Legemaate J. Verdiepingsonderzoek Uitvoering Wvggz: goede voorbeelden uit de praktijk. 2024. ZonMW, Den Haag. Beschikbaar via Eindrapport-Verdiepingsonderzoek-Uitvoering-Wvggz-def.pdf. Geraadpleegd op 9 september 2025.
29. Priebe S, Yeeles K, Bremner S, Lauber C, Eldridge S, Ashby D, David AS, O'Connell N, Forrest A, Burns T. Effectiveness of financial incentives to improve adherence to maintenance treatment with antipsychotics: cluster randomised controlled trial. BMJ. 2013 Oct 7;347:f5847. doi: 10.1136/bmj.f5847. PMID: 24100934; PMCID: PMC3805491.
30. Prinsen E, Wel B van, Mulder N, Koning N de. Handboek IHT – Intensive Home Treatment (eerste druk). 2016. De Tijdstroom, Utrecht. ISBN: 9789058983053
31. Putkonen A, Kuivalainen S, Louheranta O, Repo-Tiihonen E, Ryynänen OP, Kautiainen H, Tiihonen J. Cluster-randomized controlled trial of reducing seclusion and restraint in secured care of men with schizophrenia. Psychiatr Serv. 2013 Sep 1;64(9):850-5. doi: 10.1176/appi.ps.201200393. PMID: 23771480.
32. Ruchlewska A, Wierdsma AI, Kamperman AM, van der Gaag M, Smulders R, Roosenschoon BJ, Mulder CL. Effect of crisis plans on admissions and emergency visits: a randomized controlled trial. PLoS One. 2014 Mar 19;9(3):e91882. doi: 10.1371/journal.pone.0091882. PMID: 24647274; PMCID: PMC3960137.
33. Shea T, Dotson S, Tyree G, Ogbu-Nwobodo L, Beck S, Shtasel D. Racial and Ethnic Inequities in Inpatient Psychiatric Civil Commitment. Psychiatr Serv. 2022 Dec 1;73(12):1322-1329. doi: 10.1176/appi.ps.202100342. Epub 2022 Aug 12. PMID: 35959533.
34. Simon J, Mayer S, Łaszewska A, Rugkåsa J, Yeeles K, Burns T, Gray A. Cost and quality-of-life impacts of community treatment orders (CTOs) for patients with psychosis: economic evaluation of the OCTET trial. Soc Psychiatry Psychiatr Epidemiol. 2021 Jan;56(1):85-95. doi: 10.1007/s00127-020-01919-4. Epub 2020 Jul 27. PMID: 32719905; PMCID: PMC7847440.
35. Staring AB, Van der Gaag M, Koopmans GT, Selten JP, Van Beveren JM, Hengeveld MW, Loonen AJ, Mulder CL. Treatment adherence therapy in people with psychotic disorders: randomised controlled trial. Br J Psychiatry. 2010 Dec;197(6):448-55. doi: 10.1192/bjp.bp.110.077289. PMID: 21119150.
36. Steadman HJ, Gounis K, Dennis D, Hopper K, Roche B, Swartz M, Robbins PC. Assessing the New York City involuntary outpatient commitment pilot program. Psychiatr Serv. 2001 Mar;52(3):330-6. doi: 10.1176/appi.ps.52.3.330. PMID: 11239100.
37. Steinert T, Baumgardt J, Bechdolf A, Bühling-Schindowski F, Cole C, Flammer E, Jaeger S, Junghanss J, Kampmann M, Mahler L, Muche R, Sauter D, Vandamme A, Hirsch S. Implementation of guidelines on prevention of coercion and violence (PreVCo) in psychiatry: a multicentre randomised controlled trial. Lancet Reg Health Eur. 2023 Nov 20;35:100770. doi: 10.1016/j.lanepe.2023.100770. PMID: 38058297; PMCID: PMC10696231.
38. Swartz MS, Swanson JW, Wagner HR, Burns BJ, Hiday VA, Borum R. Can involuntary outpatient commitment reduce hospital recidivism?: Findings from a randomized trial with severely mentally ill individuals. Am J Psychiatry. 1999 Dec;156(12):1968-75. doi: 10.1176/ajp.156.12.1968. PMID: 10588412.
39. Thornicroft G, Farrelly S, Szmukler G, Birchwood M, Waheed W, Flach C, Barrett B, Byford S, Henderson C, Sutherby K, Lester H, Rose D, Dunn G, Leese M, Marshall M. Clinical outcomes of Joint Crisis Plans to reduce compulsory treatment for people with psychosis: a randomised controlled trial. Lancet. 2013 May 11;381(9878):1634-41. doi: 10.1016/S0140-6736(13)60105-1. Epub 2013 Mar 26. PMID: 23537606.
40. Tinland A, Loubière S, Mougeot F, Jouet E, Pontier M, Baumstarck K, Loundou A, Franck N, Lançon C, Auquier P; DAiP Group. Effect of Psychiatric Advance Directives Facilitated by Peer Workers on Compulsory Admission Among People With Mental Illness: A Randomized Clinical Trial. JAMA Psychiatry. 2022 Aug 1;79(8):752-759. doi: 10.1001/jamapsychiatry.2022.1627. PMID: 35662314; PMCID: PMC9171654.
41. Tseng PY, Xie XY, Hsu CC, Chien SH, Chen JD, Wang JY. Investigating Medical Cost and Mortality Among Psychiatric Patients Involuntary Admissions: A Nationwide Propensity Score-Matched Study. Psychiatry Investig. 2022 Jul;19(7):527-537. doi: 10.30773/pi.2021.0219. Epub 2022 Jul 21. PMID: 35903055; PMCID: PMC9334811.
42. Välimäki M, Lantta T, Anttila M, Vahlberg T, Normand SL, Yang M. An Evidence-Based Educational Intervention for Reducing Coercive Measures in Psychiatric Hospitals: A Randomized Clinical Trial. JAMA Netw Open. 2022 Aug 1;5(8):e2229076. doi: 10.1001/jamanetworkopen.2022.29076. PMID: 36040740; PMCID: PMC9428738.
43. van de Sande R, Nijman HL, Noorthoorn EO, Wierdsma AI, Hellendoorn E, van der Staak C, Mulder CL. Aggression and seclusion on acute psychiatric wards: effect of short-term risk assessment. Br J Psychiatry. 2011 Dec;199(6):473-8. doi: 10.1192/bjp.bp.111.095141. Epub 2011 Oct 20. PMID: 22016437.
44. Voskes Y, van Melle L, Jehoel-van As M, Mulder N, van Mierlo T. Werkboek HIC. 2026. Boom, Amsterdam. ISBN: 9789024476633.
45. Walker S, Mackay E, Barnett P, Sheridan Rains L, Leverton M, Dalton-Locke C, Trevillion K, Lloyd-Evans B, Johnson S. Clinical and social factors associated with increased risk for involuntary psychiatric hospitalisation: a systematic review, meta-analysis, and narrative synthesis. Lancet Psychiatry. 2019 Dec;6(12):1039-1053. doi: 10.1016/S2215-0366(19)30406-7. PMID: 31777340; PMCID: PMC7029280