The primary objective of the protocol is to study the long-term outcome of a large group of traumatic brain injury patients.

This outcome is to be described in terms of activity, participation, quality of life, SOCIO-professional outcome and impact on caregivers, and in relation to health care provision.

The secondary outcome is to measure the impact on functional outcome of several predictive factors, and their relative importance on outcome. Our principal hypothesis is that SOCIO-professional and health provision factors play a major role on long-term outcome, further even than initial severity of brain injury.

This work represents a part of a larger collaborative prospective study called PariS-TBI (Severe Traumatic Brain Injury in the Parisian area) financed by the French National Authority for Health. The PariS-TBI study was undertaken in 2005, to provide epidemiological data. Patients aged 15 or more were included if they had sustained a severe TBI in the Parisian area, defined by an initial Coma Glasgow Score (GCS) of 8 or less. During a 20-months period, all consecutive patients who met inclusion criteria were recruited by mobile emergency services. Data from patients' demographics, injury characteristics, acute and post acute phase were collected prospectively. Patients were evaluated through a phone interview at one year post traumatism.

A total of 504 patients were included, of which 257 were alive at the end of acute care.

Investigations of the present work will be undertaken between four and six years after brain injury for survivors. Patients of the initial cohort will be contacted twice through phone call or post. Information about this follow up study will be given orally and through written form. Informed consent of patient or legal advisor will be obtained before setting an appointment for the evaluation.

Evaluation will take place either in the patient's place of living, or in a clinical setting in the hospitals Broussais, Paris, France or Raymond POINCARE, GARCHES, France, according to the patient's choice.

Two trained neuropsychologists will lead the evaluations, during an interview with the patient and the informal care giver, defined as the person (family member or friend), most responsible for day-to-day decision making and care for the patient. If a patient refuses the interview, a short questionnaire will be proposed to him by the neuropsychologist by phone.

Demographic and pre-traumatism clinical data will be available from the initial study, as well as data on type and severity of traumatism, and on evolution during acute care. Early outcome data include cognitive disability (Dysexecutive Questionnaire), global participation (Glasgow Outcome Scale-Extended), and return to work.

Data assessed through the present study will include an evaluation of deficiencies: presence and importance of motor sequelae, cerebellar dysfunction, visual or auditive deficiencies, epilepsy, chronic pain. Cognitive deficiencies will be assessed through the Neurobehavioural Rating Scale-Revised, the Dysexecutive Questionnaire, and a questionnaire of complaints from the patient and from his informal care giver.

Activities will be evaluated by the Barthel index. Capacity of motor vehicle driving will be assessed. Participations will be measured globally with the Glasgow Outcome Scale-Extended, and specifically by a SOCIO-professional integration questionnaire and the BICRO-39 scale.

Quality of life will be assessed through the EuroQoL questionnaire, and a global evaluation of the patient's quality of life on a 0-10 numeric scale from the patient and from his care giver. The Hospital Anxiety and Depression Score will be measured.

Data on patients' pathway and intensity of clinical care will be collected: the clinical pathway from acute care to home or health care facility discharge, a quantification of clinical follow up and rehabilitation. The patient's environment (place of living, presence of informal carers) will be assessed, as well as modality of return to work.

The questionnaire will include data on financial compensations and intensity of formal home care.

Data on the quality of life of the informal care giver and an evaluation of the burden of care will be collected, using the ZARIT questionnaire.

The exploitation of this large data base will enable an up-to-date description of the long term outcomes of severe TBI patients in the Parisian area. It will be used to clarify the role of several factors on long term activities, participations and quality of life for severe TBI patients, the hypothesis being that factors directly related to the severity of the traumatism play a smaller role in long term outcome than SOCIO-environmental characteristics such as professional category, intensity of rehabilitation care, presence of informal care.

Statistical analysis will be conducted with the R® statistical software. Association between various predictive factors and outcome (GOSE, BICRO-39, EuroQoL) will be assessed through UNIVARIOUS tests, followed by multivarious regression models, using a backward selection strategy to select relevant variables.

Inclusion Criteria:

• Adults and children over 15 years old.
• Severe TBI with a Glasgow Coma Scale (GCS) score of 8 or less before admission to the hospital.
• Accident within the Regional Parisian area (Paris and 7 surrounding districts).
• Alive at emergency arrival

Exclusion Criteria:

• Dead before care by emergency

traumatic brain injury patients