Investigator's decision (departure from institution)

In 2003, MAHO study (Ferrand E, Jabre P, Vincent-Genod C, et al. Circumstances of death in hospitalized patients and nurses' perceptions: French multicenter Mort-a-l'Hôpital survey. Arch Intern Med. 2008 168: 867-875.) evaluated the way 3793 patients died in 200 French hospitals and showed that their conditions of death were not optimal. The 22th April 2005 French Law precised patient's end of life rights with necessity to refrain from any unreasonable obstinacy, the right to refuse treatments and the obligation of a collegial process decision when the patient is not conscious. Since then, studies haven't demonstrate any improvement and found that palliative strategy in France is much less used than in other developed countries.

In 2003, MAHO study evaluated the way 3793 patients died en 200 French hospitals and showed that their conditions of death were not optimal. The 22th April 2005 French Law precised patient's end of life rights with necessity to refrain from any unreasonable obstinacy, the right to refuse treatments and the obligation of a collegial process decision when the patient is not conscious. Since then, studies haven't demonstrate any improvement and found that palliative strategy in France is much less used than in other developed countries.

Principal Objective: To evaluate the impact of an early palliative strategy using vulnerability criteria compared to standard care.

Primary endpoint: Rate of withdraw/withhold of treatment in each group.

Secondary objectives: To evaluate the impact of early recognition of patients' vulnerability on death conditions ; to evaluate this strategy impact according to unit type on length of stay, palliative strategy modalities and caregivers' satisfaction.

Secondary endpoints: Rate of therapeutic involvement reflections ; rate of death following withholding or withdrawing of treatments ; traceability of the level of therapeutic involvement process ; Rate of patients deceased with their relatives next to them ; rate of patients deceased with comfort treatment ; rate of palliative care consultation before death ; rate of asks for euthanasia ; Doctor and nurse's perception of quality of support and death process of the patient

Methods: Prospective, controlled, cluster randomized study of routine care 2 groups:

• Group A: standard care and practice after 1 day of training
• Group B: 1 day of training, learning the vulnerability criteria that should induce early thinking about level of therapeutic involvement; web accessed forms will be available to help collegial process, withhold and withdraw decisions traceability, using legal requirements Number of patients to include: To detect a 20% absolute difference in palliative strategy used (30 to 50%), we determined that 5040 patients would provide a power of 80% with the use of a two-sided alpha level of 0.05. Sequential analysis is planned in order to early stop the study in case of efficacy or futility (minimal inclusion: 500 patients)

Inclusion criteria:

All patients hospitalized with at least one of the following vulnerability criteria will be included:

• Evolutive and symptomatic incurable cancer
• Aged more than 75 years old and presenting several geriatric syndromes (cognitive disorders, isolation, malnutrition, bedridden more than 12h per day)
• Neurologic pathology, chronic, with loss of autonomy (Performance Status>3)
• Final organ failure (heart, lungs, liver, kidney) with loss of autonomy (Performance Status>3)
• Care refusal and/or expressed will to die or repeated request for help to die

Exclusion criteria:

• Minors
• Patients without indication for treatment or surveillance with length of stay inferior to 24h
• Brain dead patients
• Not consent patients

Duration: 37 months (28 inclusion months for each center and a follow-up to hospital discharge, death or 9 months if the patient is still hospitalized).

Number of participating centers: 20 centers (28 services) were selected and recruit after training program among centers that did not used a formalized process to initiate level of therapeutic involvement reflection.

Inclusion Criteria:

• Patient more than 18 years old hospitalized for at least 24h and who's prognosis (survival or quality of life) should lead to a palliative approach

• Patient suffering with at least one of the following vulnerability criteria will be included:

  • Evolutive and symptomatic incurable cancer
  • Aged more than 75 years old and presenting several geriatric syndromes (cognitive disorders, isolation, malnutrition, bedridden more than 12h per day)
  • Neurologic pathology, chronic, with loss of autonomy (Performance Status>3)
  • Final organ failure (heart, lungs, liver, kidney) with loss of autonomy (Performance Status>3)
  • Care refusal and/or expressed will to die or repeated request for help to die

• No opposition to the use of data collected from the patient or a relative or inclusion in emergency and non-opposition collected offline

Exclusion Criteria:

• Minors
• Patients without indication for treatment or surveillance with length of stay inferior to 24h
• Brain dead patients
• Not consent patients