Objective – The QUALYOP study has three objectives: 1) To describe both structural resources and organizational procedures of the 12 Psychiatric Hospitals and to assess the rate of decline in the number of in-patients; 2) to assess the quality of structures, organizational procedures and activities delivered by the hospitals; 3) to describe the socio-demographic, clinical and disability characteristics of the in-patient population. Data regarding this third objective will be presented in a future paper. Design – A transectional design has been adopted. The 12 Psychiatric Hospitals have been visited between July and November 1994. Administrative informations have been collected using an ‘Hospital Form’, filled up by the director of each Hospital; an ‘Hospital Ward Form’, filled up by a group of trained researchers, was adopted to evaluate the quality of each Hospital Ward. A reportage of the structural aspects of the hospitals has also been performed by a professional photographer. Setting – The 12 former Psychiatric Hospitals of Regione Lombardia (Bergamo, Brescia, Castiglione delle Stiviere, Codogno, Como, Cremona, Limbiate, Mantova, Milano, Sondrio, Varese, Voghera). Main outcome measures ‖ We utilized a set of indicators to evaluate the following three dimensions: human and logistic resources, organizational procedures, psychosocial rehabilitation activities. The quality of care has been judged in comparison with explicit reference criteria. Results ‖ 2752 patients are sheltered in the 12 Psychiatric Hospitals of Regione Lombardia. The quality assessment phase of the study has revealed that structures are inadequate in the majority of the wards (70%). In addition, in more than 70% of the wards procedures and rehabilitation activities aiming at decreasing the disability of patients have not been activated. Conclusions ‖ A low degree of consistency between avaliable structures and quality standards has been evidentiated. In addition, soft indicators have documented the inadequacy of organization procedures and activities. These results demonstrated that the dehospitalization policies are often inadequate and that educational policies are warranted to train and motivate mental health professionals. It has been also demonstrated the feasibility of Quality Assurance studies which use a system of reference criteria derived from reliable clinical studies, ethical values, experience, definition of ‘not admissible events’ and common sense.