An Evidence-Based Educational Intervention for Reducing Coercive Measures in Psychiatric Hospitals

Key Points Question Does an evidence-based educational intervention for nurses decrease the use of seclusion rooms in psychiatric hospitals compared with usual practice? Findings In this cluster randomized clinical trial of 8349 patients receiving care in 27 wards at 15 psychiatric hospitals across Finland, the use of a seclusion room at the ward level decreased from 15.1% to 14.3% in the intervention group and increased from 13.9% to 18.7% in the usual practice group. Meaning Although feasible, the educational intervention had a limited effect for change of occurrence rate of patient seclusion.


eAppendix 1. Differences between the protocol and trial
In the protocol, the primary outcome was incidence of seclusion room use by individual patients. In the trial, however, we were not able to calculate incidence rate at the patient level due to a lack of information on individual patients for each treatment period. Hence, we evaluated the occurrence rate of seclusion use on ward level. Our sensitivity analysis described in the Results suggest that the key finding in the primary outcome was robust and not affected by the change in data analysis plan.
We would have preferred to analyse the data from three time points (2015,2016,2017). However, the data were analysed at two time points only (baseline vs. post-allocation) because the intervention periods partially overlapped in year 2016.
Another set of analysis was conducted to take into account a possible hospital effect on study results.

eTable 1. Description of the evidence-based intervention pathway Phases
What happened in the intervention?

Acceptance
• To ensure acceptance of the changes, possible problems in current treatment practices in each ward identified by analysing the current ward 'rules'; the quality of the service facilities assessed from the point of views of patients, families, and staff members. • Opinions about the possible changes discussed and the quality gaps in services identified. • Areas to be developed were identified in consultation with staff members in two educational seminars (8/2016, 11/2016) • The strengths, weaknesses, opportunities, and threats (SWOT) related to future changes identified.

Applicability
• The facilitators visited the intervention wards two times times during active education phase (5-6/2016, 12/2016-1/2017) and later in maintaining phase (10/2017) to support identification of ward-specific quality gaps, and discussing with ward managers, patients, relatives and staff members at local meetings. • Barriers and facilitating factors for change identified.
• Decisions made regarding what development activities to be done and how

Available
• The evidence-based part supported by learning material to ensure staff's competence to make managed changes regarding treatment practices.

Able
• The facilitators made monthly calls to on each ward to prompt, encourage and monitor new treatment practices (1-9/2017). • Ongoing identification of areas to be developed were discussed in one seminar (5/2017). • Continuity of identification of ward-specific quality gaps, and discussing with ward managers, patients, relatives and staff members at local meetings was supported in seminar (10/2017) • Possible problems in changing practices shared, and solutions further explored. • Project leaflets were distributed to encourage staff members and disseminate progress and achievement of the study.

Acted on
• The facilitators offered hands-on support to study ward to ensure that wards were acting based on the specific implementation plan (problems identified on the ward, a new solution found for their problems, maintaining new practices). • Each ward monitored in their progress using a specific monitoring form as evidence of the changes in clinical practice.

Phase 6
Agreed on • As evidence of changes in daily practices, the content of the monitoring forms and local house rules re-analysed. • Changes in treatment practices were recorded by comparing the situation at the beginning and at the end of the project.

Adhered
• Staff adherence of the educational process was evaluated by assessing how specific intervention fidelity criteria had been fulfilled. -There was no tools in the comfort room for patients to use, to calm down 6 -A room that is safe and where you can go to calm down will be created in the ward -Seclusion room will be used for calming down -Comfort rooms will be equipped with e.g. a weighted blanket, stress ball, rubics cube and other tools to reduce anxiety -Permission is not given to acquire the furniture for comfort room or there are no suitable furniture existing in the storage.
-Comfort room will be used for recreational purposes by patients e.g. to lie and listed to music loud or for self-harm. Patients might therefore start to behave poorly to get to the comfort room -In some wards, frequency of using the room has been low.
-When comfort room is used, reduces conflicts, selfharm, use of medication and frequency and length of seclusion and physical restrictions, and helps reducing anxiety, agitation and insecurity of patients'

eAppendix 3. Evidence existing prior to this study
Staff education for the reduction of coercive measures is widely supported by professional associations, but fewer evidence-based and tested educational protocols supporting optimising outcomes are available. On-the-job training is widely used in psychiatric hospitals although its effectiveness has seldom been evaluated with randomised controlled trials. We designed the study in 2017, when no systematic reviews had been published, and only a few evaluations of the effectiveness of non-pharmacological studies to reduce the use of coercive measures had been published, to our knowledge. In our large-scale study, we aimed to design an educational intervention using an evidence-based pathway for how to make changes using systematic One study 3 included self-reflective exercises in their intervention to help staff members to manage patients' challenging behaviour on the wards and showed that the three-month intervention reduced the lengths of patient mechanical restraints during the seclusion period.
Two studies 1,4 used structured risk assessment methods to assess patient violence on daily basis.
As an outcome, one study 4 showed a significant decrease in coercive measures used, and another study 1 showed a significant decrease in the lengths of patient seclusions. One study 2 used multiple methods to combine post-incident analysis and a review of the problems and coercion incidences, staff counselling and discussions, and tailored crisis plans for patients. This one-year intervention was found to be effective in reducing days where seclusion, restraint and room observations occurred, and decreasing the lengths of seclusion and restraint. In addition, one intervention 5 included a package of ten interventions used in a three-month period (e.g., standards of behaviour, a de-escalation model for staff, sharing good things about each patient, inter-patient support meetings, distraction and sensory modulation tools to use with agitated patients). The intervention was effective in reducing the use of coercive measures.
However, all five studies employed a cluster randomized design, but failed to include cluster effects in their estimates.

Added value of this study
Our study tested an educational intervention where an evidence-based pathway was used to decrease the use of seclusion rooms in psychiatric hospitals. This process was led by the research personnel, but changes implemented on study wards were implemented by nurses. The educational intervention combined evidence-based knowledge shared with nurses, user-centred approaches focusing on the needs of patients, family members, and nurses based on lived experiences in clinical practice. Responses to these problems were solved using evidence-based, tailored treatment methods. The educational intervention with multiple components had only a weak effect in reducing the use of seclusion rooms on psychiatric wards. At the same time, the use of forced medication increased partially. Despite the strong engagement of various stakeholders in the study, no changes were found in nurses' team climate or patient-related outcomes.

Implications of all the available evidence
Considering all the available evidence, multicomponent educational interventions for nurses could reduce the use of seclusion rooms and the number of patient restrictions in psychiatric hospitals, but other form of coercion, such as the use of forced medication may concurrently increase. These multicomponent educational packages with more focused intervention elements toward coercion practices could have some potential to shift practice and treatment culture toward less coercive care at least in Europe. Whether this is evident in middle-and low-income countries should be tested more widely. Therefore, the results of this trial have opened the doors to very large trials across the world. In addition, as the available evidence today focuses on shortterm outcomes, future research is needed to test the best combinations of intervention components to reduce all types of coercion with lasting effects.