The systematic review by Papoulias et al ^{Reference Papoulias, Csipke, Rose, McKellar and Wykes1} on the psychiatric ward as a therapeutic space reminds us of the important effects of environmental factors on in-patients. The physical environment is likely, however, to be particularly significant in settings where length of stay is long, whether in or out of hospital. The 1995 book by Halpern ^{Reference Jenkins, Dye and Foy2} describes the mental health effects of the built environment on residents of a housing estate, and the concerns of Papoulias et al should be explored in residential mental health facilities in the community.

For psychiatric in-patients, patient characteristics (including diagnosis) and psychosocial environmental factors are powerful determinants of what happens in the hospital, including behaviour disturbances, service user opinions and also, sometimes, illness outcomes. ^{Reference Bowers, Brennan, Ransom, Winship and Theodoridou3,Reference Dix, Pereira, Chaudry, Dale and Halliwell4} Clark ^{Reference Ulrich, Bogren and Lundin5} was one of those who showed that different wards for different varieties of patient should have different sorts of environment, drawing on the extensive previous research in this field (e.g. Stanton & Schwartz⁶). A major problem with today’s in-patient wards is that everyone has to be admitted to, and as like as not, stay in, the same environment, whether or not it suits them and their illness. This might remind clinicians with long memories of the features of the old observations wards, to which anyone putatively mentally ill could be admitted, primarily for triage and transfer to the setting which suited them best. Today there is, in these terms, only the triage.