The purpose of the study is to compare the effectiveness of Divomed in organising complex hospital discharges with a conventional organisation. Effectiveness will be assessed by reducing the length of stay of patients in geriatric short-stay care.

This is an interventional, randomised (ratio 1:1), single-centre, open-label study evaluating Divomed, a digital tool combined with human coordination, to speed up the discharge of people aged over 75 from unscheduled short-stay geriatric hospital admissions to their homes.

The study population will be patients aged 75 or over, admitted for short term unscheduled geriatric hospitalisation and eligible for discharge home.

Inclusion Criteria:

• Age ≥ 75 years ;
• Unplanned hospitalisation from emergency, transfers from other inpatient departments or outpatient medicine;
• Patient eligible for short-stay geriatric hospitalisation estimated to last at least 10 days;
• Patient eligible for a return home;
• Patients who have been informed and have signed an informed consent form. If necessary (for cognitive or other reasons), the information must be given to a third party, i.e. the patient's legal representative or trusted person (independent of the investigator and the sponsor), who may be asked to sign or co-sign the informed consent form (depending on the patient's vulnerability).

Exclusion Criteria:

• Patients scheduled for transfusion, iron infusion or colonoscopy;
• Patients in palliative care;
• Geographical area other than department 37;
• Planned discharge from hospital to a residential establishment for dependent elderly people or to follow-up and rehabilitation care or long-term care units;
• Patient or deprived of liberty by judicial or administrative decision ;
• Patient participating in or being excluded from another clinical trial;
• Patients not covered by a social security scheme.

open-label study