The literature reports many changes in mental hospital philosophy and attitudes that reflect on security practices. Since these practices appear to have a direct bearing on requirements for physical facilities, the Architectural Study Project sought to (a) assess the extent and degree of privilege and responsibility permitted to patients in the various types of mental hospitals in this country (b) to establish a base line in reference to security practices in order to detect future changes and trends; and (c) to identify some of the factors which might be related to security practices in mental hospitals.[Refer Source Pdf for the Table 5]A mail questionnaire sent to all public mental hospitals in the United States, Canada, Hawaii and Puerto Rico, resulted in information from 94% of the hospitals which also contained about 94% of all beds in all public mental hospitals.Analysis of the data showed that 12.8% of all the patients in all the hospitals are on open wards, 22.3% are privileged, 45.8% are semi-privileged and 31.9% are non-privileged.State, VA, county, Canadian and Hawaiian hospitals appear to differ significantly in their patient privileges practices. VA hospitals appear to have the smallest proportion of non-privileged patients and semi-privileged patients. Except for the one Hawaiian hospital, the VA hospitals also have the largest percentage of privileged patients with county hospitals in second place. Hospitals which are primarily or entirely for the criminally insane have the lowest privilege rate. Canadian and state hospitals are fairly similar in their practices.Variation in patient privilege practices within each hospital group seems to be related to size of the hospital. Doctor-patient ratios and recency of hospitals do not seem to be related to privilege practices except in VA hospitals where hospitals established after 1946 showed higher percentages of open ward patients than those established prior to this date. Geriatric, medical-surgical and tuberculosis services show a generally lower privilege level than the rest of the services in mental hospitals.The trend of present day psychiatric thinking and practice is toward the open hospital. Yet the data presented here shows that there is still a very long way to go. It would be of utmost importance to determine those factors that bear upon this issue. This study did not attempt to discover the role of staff attitudes in security practices but evidence is available (Greenblatt (2), Stanton (6), our study (4)) that this may be a key factor.