[Description of various types of intensive care and continuous monitoring in France].

There are multiple types of intensive care units (ICU), and the purpose of this study was to describe the activity of different ICUs, using the same method. Thirty-eight ICUs were chosen by co-optation rather than by randomization. The data collected concerned inputs (patients), processes (treatments) and outcomes (ICU death rate). The ICUs were divided into three intensive care groups (ICG): M = medical, C = surgical and P = surgical, scheduled or elective. The 3,687 patients studied were classified as: M = 2,175, C = 885 and P = 627. The first part of the results concerned differences between the three ICGs: inputs, processes and outcomes were very different in the M, C and P groups, particularly in the P (elective) group where the therapeutic level was highest for low simplified acute physiology scores (SAPS) and the mortality was lowest for high SAPSs. The second part of the results concerned differences between ICUs. Intermediate units had older, less severe and mainly medical patients. Surgical patients were younger, had a better previous health status and 40 per cent of them were scheduled. Therapeutic intervention scoring system (TISS) points were higher, mainly due to a higher rate of ventilated patients and of patients fitted with pulmonary artery lines on the first day. It is concluded that different intensive care units can be described with the same tools.