‘Open door policy’ refers to a policy of maintaining open doors in mental health settings and particularly hospital-based settings that otherwise would be ‘closed’ or ‘locked’. Germany appears to have the most advanced use and evaluation of open door policies in acute psychiatric settings in high-income countries (Gooding et al., 2020, p.33). A trial has also occurred in Switzerland, at the Universitäre Psychiatrische Kliniken (UPK), and in the UK, efforts have been undertaken to canvas the views of stakeholders on open door policies (Bowers et al., 2010).

Other services discussed in this report are likely to practice some form of ‘open door policy’, even as they may not describe it this way (see, for example, the Norwegian BET initiative above). In Italy, for example, Roberto Mezzina (2014, p.440) reports on the outcomes of an ‘open door... no restraint system of care for recovery and citizenship' in the city of Trieste, Italy. The Trieste model is discussed below in the Hybrid Approaches Section. Other mental health crisis services not included in this report may also practice some form of an open-door approach.

A compilation of research on the use of an ‘open door policy’ or ‘open acute psychiatry’ in mental healthcare settings in Germany was submitted to the DH-BIO. This included material from two implementation sites (UPK in Basel and Charité, Universitätsmedizin Berlin) as well as supplementary materials concerning its use in 21 German hospitals (Huber et al., 2016a; see also Cibis et al., 2017; Huber et al., 2016b; Lang et al., 2016; Lang & Heinz, 2010; Schneeberger et al., 2017).

Christian Huber and colleagues (2016a, 2016b), and Andres Schneeberger and colleagues (2017) undertook two largescale studies of service data concerning 349,574 admissions to 21 German psychiatric inpatient hospitals from 1998 to 2012. Huber and colleagues (2016a) sought to compare hospitals with and without ‘locked wards’. They reported that treatment on ‘open wards’ was associated with a decreased probability of:

• suicide attempts,
• absconding with return, and
• and absconding without return.

Treatment on an open ward was not associated with a decreased probability of death by suicide (Huber et al, 2016a). In a second study using the same dataset, Schneeberger and colleagues measured the effects of open versus locked door policies against rates of ‘aggressive incidents’ and restraint/seclusion and found that both aggressive behavior and ‘[r]estraint or seclusion during treatment [were] less likely in hospitals with an open door policy’ (Schneeberger et al., 2017).

Some concerns have been raised that the term ‘open door policy’ was classified arbitrarily in the data set (Pollmächer and Steinert, 2016). Huber and colleagues (2016b. p.1103) refute this claim, and conclude as follows:

With respect to patient safety and coercive measures, results of previous studies have observed that opening formerly locked wards reduces violence and coercion. This reduction does not happen at the expense of placing aggressive patients on other still locked wards, and is not compensated by an increase of coercive measures elsewhere. In our experience, a change in professional attitudes towards patients that promote personal contact and de-escalation is a key component of open door policies. (Huber et al., 2016b. p.1104) (original citations removed)

It is outside the scope of this compendium to engage with debates about German open door policies, though they should be considered by actors seeking to replicate the practice.

Another German study published in 2017, by Mara-Lena Cibis and colleagues (2017), compared the impacts on an acute psychiatric hospital ward of phases in which the ward was ‘closed’ and ‘90% of daytime opened’. The authors observed that during the phase of opened doors there were ‘significantly reduced aggressive assaults (p < 0,001) and coercive medication (p = 0,006) compared to the closed setting, while the absconding rate did not change (p = 0,20)’ (Cibis et al., 2017, p.141). The authors noted a limitation that the ‘retrospective non-experimental design’ means that ‘no causal interpretations can be drawn’ but concluded that the results support the claim that open door policies are ‘associated with reduction of aggressive assaults and coercive medication without increasing absconding rates’ (Cibis et al., 2017, p.141).

The research on open door policies suggests that factors which facilitated its implementation include strong psychotherapeutic skills among staff, highly motivated and well-staffed nursing personnel, and the commitment of ward and hospital leadership, who must take responsibility for the initiative.

Service users report greater satisfaction with the practice. The initiative, according to the submission, results in decreased appeals against treatment, decreased discharges against medical advice, decreased coercive measures, enhanced ward atmosphere, and decreased transfer of patients.