Admitting a very elderly patient to the Intensive Care Unit (ICU) is one of the most difficult clinical challenges in medicine. There are few data to help guide clinicians in this area: estimates of the benefits of ICU admission, especially in the very elderly, are sparse. Rates of ICU admission of very elderly thus vary widely by hospitals.

The ICE-CUB1 (PHRC AOR 03 035) project has studied the ICU admission decision process of patients over 80 arriving in Emergency Departments (ED) with conditions that potentially warrant ICU admission and their outcome six months after ED visit. Overall rate of patients deemed eligible for ICU admission was of 12% (Garrouste et al. Crit Care Med 2009) but ranged from 5% to 38% across the participating centers. This variability persisted after adjustment for patients' characteristics (MOR 2.25, 1.60-3.58; hospital-related variance 18%). The analysis also revealed that high functional status prior to ICU visit, good nutritional status as assessed by an emergency physician and the absence of cancer were of good prognosis for outcome six months after ED visit. Only 23% of patients visiting the ED with a life-threatening condition and all positive prognostic factors mentioned above were admitted to an ICU in the ICE-CUB1 study.

Hypothesis Elderly patients visiting the ED with a life-threatening condition, high functional status prior to ICU visit, good nutritional status as assessed by an emergency physician and no cancer will potentially benefit from ICU care and should thus be admitted.

Main objective Determine whether a strategy consisting of recommendations of ICU admission of all patients over 75 visiting the ED with a life threatening condition, no cancer, good functional and nutritional status prior to ED visit decreases the mortality of these patients six months after ED visit.

Secondary objective : Assess the impact of the strategy on:

• In-hospital mortality
• Rate of ICU admission
• Place of living and quality of life six months after ED visit

Primary outcome :Mortality six months after ED visit Secondary outcomes

• In-hospital mortality
• ICU admission
• Change in functional status six months after ED visit
• institutionalization
• Quality of life six months after ED visit

Admitting a very elderly patient to the Intensive Care Unit (ICU) is one of the most difficult clinical challenges in medicine. There are few data to help guide clinicians in this area: estimates of the benefits of ICU admission, especially in the very elderly, are sparse. Rates of ICU admission of very elderly thus vary widely by hospitals.

The ICE-CUB1 (PHRC AOR 03 035) project has studied the ICU admission decision process of patients over 80 arriving in Emergency Departments (ED) with conditions that potentially warrant ICU admission and their outcome six months after ED visit. Overall rate of patients deemed eligible for ICU admission was of 12% (Garrouste et al. Crit Care Med 2009) but ranged from 5% to 38% across the participating centers. This variability persisted after adjustment for patients' characteristics (MOR 2.25, 1.60-3.58; hospital-related variance 18%). The analysis also revealed that high functional status prior to ICU visit, good nutritional status as assessed by an emergency physician and the absence of cancer were of good prognosis for outcome six months after ED visit. Only 23% of patients visiting the ED with a life-threatening condition and all positive prognostic factors mentioned above were admitted to an ICU in the ICE-CUB1 study.

Hypothesis Elderly patients visiting the ED with a life-threatening condition, high functional status prior to ICU visit, good nutritional status as assessed by an emergency physician and no cancer will potentially benefit from ICU care and should thus be admitted.

Main objective Determine whether a strategy consisting of recommendations of ICU admission of all patients over 75 visiting the ED with a life threatening condition, no cancer, good functional and nutritional status prior to ED visit decreases the mortality of these patients six months after ED visit.

Secondary objective : Assess the impact of the strategy on:

• In-hospital mortality
• Rate of ICU admission
• Place of living and quality of life six months after ED visit

Primary outcome :Mortality six months after ED visit Secondary outcomes

• In-hospital mortality
• ICU admission
• Change in functional status six months after ED visit
• institutionalization
• Quality of life six months after ED visit

Type of study Cluster stratified randomized controlled trial. Stratification criteria are existence of an acute geriatric ward, capacity of the emergency department and location of the hospital (in or out Paris area)

Inclusion Criteria:

• Age over 75 years old
• At least one organ failure
• No cachexia
• No active known cancer
• Good functional status (as assessed by an ADL score > 4) or not evaluable
• Affiliated to social security

Exclusion Criteria:

• refusal

Patients over 75 visiting the emergency department with a condition potentially warranting intensive care unit admission