Long-term outcome after head injury

Bandolier often wonders about the natural history of disease or disorder. Observational studies which give us answers are hard to find, but the long-term outcomes after head injury have been described in a good study from France [1].

Trauma and its sequelae are major health problems, and it is one of the major if not the major cause of death in children and young adults. Head injury has a reported incidence of 150 to 300 per 100,000 and is one of the most frequent injuries in trauma.


The study was conducted in Aquitaine (population 2.7 million), using data from a 1986 population-based study looking at all injuries serious enough to result in death or hospital admission. Out of the 7,281 patients identified, a cohort of 1,005 which included 407 head injury patients was selected for follow up (selection criteria were not given other than that three centres were involved).
Head injury was defined as loss of consciousness, abnormal neurological examination, abnormalities on computerised tomography, or skull fractures. Trauma severity was rated using the Abbreviated Injury Scale (AIS) :
  • AIS 1-2 short period of unconsciousness or linear skull fracture
  • AIS 3 complex skull fracture or cerebral contusion
  • AIS 4 prolonged unconsciousness or intracranial haematoma
  • AIS 5 unconsciousness >24 hours, diffuse brain lesions, or severe mass effect.
Survivors were sent a letter five years after the initial trauma and were interviewed with a 200 item questionnaire at the institution or at home, or by phone, or by filling in the questionnaire at home and sending it back by post. Wherever possible a close family member was interviewed separately to supply information on the patient's behaviour. Out of the 407 head injury patients, 64 had died, 36 were lost to follow up, and three refused to participate, so data were available on 304 patients.


Patients in the cohort were predominantly under the age of 60 years (90%), and 50-60% were under 30 years at the time of their injury. About two thirds were male, and almost all had received their head injury in a traffic accident or fall. Almost all the patients lived at home. About 4% needed family support because of behavioural or cognitive problems, and this was permanent in about 20% of the most severe cases. Epilepsy and hemiparesis were present in 0.7% of the whole head injury cohort.

But overall outcome was good. Using the Glasgow Outcome Score, recovery in survivors was good in 98, 97, 88 and 47% of patients in AIS 1-2, 3, 5 and 5 respectively.
An impairment was defined as any loss or abnormality in body structure or function. Headaches, dizziness, memory disturbance, depressive mood, irritability and anxiety were common. They occurred in 25% or more of patients of all AIS classes, and there was a tendency for some of these to be more severe in those in AIS 5.

A disability was defined as a change in ability to perform an activity within the range considered normal. Few patients in AIS 2-3 were affected, but disability was higher in AIS classes 4 and 5, with difficulty in walking, taking public transport, dressing, washing and dealing with paperwork.
A handicap was defined as a disadvantage which limited the fulfilment of a role that was normal for that individual. At the time of their injury 37% of patients were working. After five years 15% could not work any more because of their injury.

Ninety-six patients were children at school in 1986. It was estimated that 7% of these students had problems in studying again, mostly because of behavioural changes.

Death rates were class dependent. They were 5, 8, 19 and 56% in AIS 1-2, 3, 5 and 5 respectively.


Bandolier liked this paper because it gave information of the likely five year outcome following a head injury. The study may be criticised because it looked at a cohort, rather than the whole head injury population of the region. The datum point was people who had a head injury as long ago as 1986, and it is possible that management has improved since then.

But it probably gives a fair impression of what can be expected in terms of recovery and problems for someone having a head injury. Useful for doctors, patients and families.


  1. F Masson, J Vecsey, LR Salmi et al. Disability and handicap 5 years after a head injury: a population-based study. Journal of Clinical Epidemiology 1997 50: 595-601.

previous or next story in this issue