[Toward a medical approach to the evaluation of activity in a physician-staffed mobile intensive care unit]

To test the ability of various medical criteria for classifying the patients in a physician-staffed mobile intensive care unit (MICU) by referring to intervention times.

Prospective, open study.

For all the on-scene interventions of the MICUs over a 10-month period, the following data were prospectively collected: pre-hospital diagnosis, initial severity score, medical care score, immediate outcome and three intervention times: on-scene time (OS), time spent with the patient by the MICU team (MT), total duration of intervention (TD).

A total of 3,672 MICU interventions were included. Median times were 45 min (32-59) for OS, 66 min (41-91) for MT and 85 min (61-116) for TD. The amount of interventions in a city was correlated with the population (R = 0.95; P < 0.001). The medical care score was greater than one in more than half of the patients. It defined five groups of patients which were different for the three intervention times (P < or = 0.001). A third of the patients were directly transported by the MICU to an ICU. For the median test, immediate outcome groups were different for the three intervention times (P < 0.001). After exclusion of patients with initial cardiac arrest, initial severity score defined five groups of patients which were different for the three intervention times (P < 0.002). Initial severity score and medical care score were correlated (R = 0.37; P < 0.001).

A classification of the patients based on immediate outcome would be a more accurate indicator of the variability in medical care and consumption of resources in a physician-staffed MICU. In addition, a medical intervention score should be developed to better characterise this medical activity.