In reply [to Schneeberger, Huber and Lang]

We thank Schneeberger and colleagues for their kind remarks on our study. They raise 2 issues concerning this article.
First, they point to the fact that we limited our inclusions to randomized clinical trials (RCTs) to obtain the highest level of evidence, thus excluding a large part of the literature. We acknowledge that quasiexperimental and naturalistic studies may be valuable. However, RCT-derived evidence on a certain intervention will always prevail over evidence derived from quasiexperimental and naturalistic studies. This has been clearly pointed out in the discussion on community treatment orders.
Second, Schneeberger et al state that inpatient interventions, such as open-door policies in psychiatric wards, lead to a reduction of compulsory measures, referring to their own research. An admission is just one point in a patient’s trajectory through mental health care, and it is an appealing idea that reducing coercion along the total mental health chain, including ward stays, might lead to less compulsory measures.
[...]