Brain Advance Access originally published online on August 8, 2007
Brain 2007 130(10):2520-2527; doi:10.1093/brain/awm185
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Death rate is increased for at least 7 years after head injury: a prospective study
1Psychological Medicine, University of Glasgow, Gartnavel Royal Hospital, 1055 Great Western Road, Glasgow G12 OXH, UK and 2NHS Quality Improvement Scotland, Delta House 50 West Nile St, Glasgow G1 2 NP and Division of Clinical Neurosciences, University of Glasgow, UK
Correspondence to: Professor Tom McMillan, Psychological Medicine, University of Glasgow, Gartnavel Royal Hospital, 1055 Great Western Road, Glasgow G12 OXH, UK E-mail: t.m.mcmillan{at}clinmed.gla.ac.uk
 |
Summary |
|---|
 Top Summary IntroductionMethodResultsDiscussionReferences |
|---|
Although a high mortality rate among patients recently admitted to hospital with severe head injury is well recognized, less is known about their later mortality and very little about the ensuing lifestyle and less-severe injuries that lead to death. The aims of this study were to determine the rate of death in the first and six subsequent years after head injury, in a prospectively identified cohort admitted to hospital, and investigate the factors associated with death—comparing these with general death rates in Scottish populations. A structured sample of 767 patients aged 14 years and over was identified at the time of admission to hospital after a head injury and followed up 7 years later. A trace exercise was conducted to identify those deceased. The General Register of Scotland confirmed death and provided information about cause of death. Seven years after head injury, 206/767 (27%) people had died. Compared to the Glasgow population, risk of death was high after head injury in months 1–2 (23 times), 3–12 (3 times) and 13–84 (2 times), and overall was especially raised in younger people, even late (13–84 months) after injury (7 times). Mortality was only associated with greater severity of head injury during year 1. Pre-injury medical history was associated both with earlier and later deaths, but risk of death remained higher in those with no such history. Later deaths were often associated with lifestyle post-injury. The primary causes of death after head injury were the same as those in the general population. Compared to the general population, the death rate after admission to hospital with head injury remains high for at least 7 years, and is particularly high for those aged under 55 years. Interventions aimed at change in lifestyle may reduce this continuing excess mortality.
Key Words: head; brain; injury mortality; outcome
Abbreviations: GROS, General Register Office for Scotland
Received March 27, 2007. Revised July 13, 2007. Accepted July 16, 2007.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodResultsDiscussionReferences |
|---|
Injuries are the leading cause of death in adults and head injury accounts for the majority of accident deaths (Tagliaferri et al., 2006
). A high early mortality among those admitted to hospital with a severe head injury is well recognized (Gennarelli et al., 1989; Sosin et al., 1995; Jennett, 1996; Ritchie et al., 2000; Mosenthal et al., 2002; Susman et al., 2002; Coimbra et al., 2003; Brown et al., 2004); much less is known about later mortality and very little about events after injuries of lesser severity, with many studies having important methodological shortcomings (Conroy and Kraus, 1988; Tagliaferri et al., 2006). Previous work has largely focussed on death during hospital admission or follow-up within the first year after severe head injury. Although thereafter, the death rate in survivors is much lower, these survivors are characteristically young (e.g. median age 39 years; Thornhill et al., 2000) and later mortality has been reported to continue to be raised (Shavelle et al., 2001; Pentland et al., 2004), and also not to differ from expected (Brown et al., 2004). Where late mortality has been observed after mild and severe injuries, postulated connections have included reduced mobility, suicide and ‘lifestyle’ (Shavelle et al., 2001; Teasdale and Engberg, 2001; Pentland et al., 2004). The purpose of the present study was to define the early and late mortality rates following head injury in a prospective, structured sample, to investigate the characteristics of deceased patients, to examine causes of their death and to compare the findings with the death rates in both the local Glasgow and Scottish populations.
|
Method |
|---|
|
TopSummaryIntroductionMethodResultsDiscussionReferences |
|---|
Participants
The original cohort, the ‘Hospitalised Head Injury Study’, was recruited prospectively and consisted of 2995 people aged 14 years and over, identified contemporaneously as admitted to a hospital in Glasgow with a clinical diagnosis of acute head injury between February 1995 and February 1996. Head injuries were identified through frequent visits by researchers to each admitting hospital and by close co-operation with hospital staff. Head injury severity was defined using the Glasgow Coma Scale. All who had a severe injury (GCS = 8 or less; n = 101) or a moderate injury (GCS = 9–12; n = 133), a random sample of those with a mild/minor injury (GCS = 13–15; n = 507) and 28 whose severity was unclassified (n = 28) were selected for follow up at 1 year and comprise the sample of 769 (Thornhill et al., 2000
). In the present study, those from the original 769 who died within 7 years of injury were compared with survivors and with data for the Scottish population.
Design
A retrospective analysis of cause of death information provided by the General Register Office for Scotland (GROS), death rates, demographic and other information collected at time of entry to the original study and at 1-year follow-up. Deaths after head injury are considered in the context of the death rate in the general population from which they come. Glasgow has higher rates of chronic ill-health and social deprivation than other cities in Scotland (McLoone and Body, 1994). Hence comparisons between deaths after head injury in Glasgow are made with expected deaths in the Glasgow population, and also for Scotland overall.
Procedure
Identifying the deceased
Of the 769 people in the sample, two had no date of birth recorded, were not traceable in the original study and were excluded from consideration here as they could not be matched by GROS (see later). At 1-year follow-up 74 were reported to have died (Thornhill et al., 2000). The names and dates of birth of the remaining 693 were given to Greater Glasgow Health Board Patient Registration. They confirmed the General Practitioner's identity using the Community Health Index. This allowed each person to be allocated as registered with a GP, as deceased (and date of death provided), not registered with a GP, as having moved to another Health Board or as not traced. The Health Board's data confirmed that 503 were alive. A list of the remaining 190 plus the 74 reported to have died before 1 year was sent to GROS for confirmation of the death and for information about the cause of death.
The GROS matches people according to sex, surname, date of birth and postal code information. They categorize their matches as Good (sex, surname, date of birth and postcode all match), Possible (sex, surname, plus date of birth or postcode information match), No Match or Not Listed. The GROS returned with Good Match, Possible Match and No Match information. The total number of 206 classified here as deceased are those who were identified as Good Match and Possible Match. The remaining 58 were returned as No Match.
Cause of death
The GROS code death according to the International Statistical Classification of Diseases, Injuries and Causes of Death. Information gathered from the medical certificate in addition to any further information provided by the certifying doctor is used to classify the underlying cause of death. Deaths occurring in Scotland after January 1st, 2000 are coded using the tenth revision (ICD-10) and deaths before this date are coded using the ninth revision (ICD-9). The ICD groups causes of death into 20 main categories, which are then further subcategorized. For simplicity we coded causes of death using the same 20 main categories of death. Only the primary cause of death for each patient was analysed.
Death rates
Death rates were constructed for the primary causes of death. To allow comparison with published statistics for the Scottish population, death rates were also matched for age and sex. Age groupings were chosen (15–34, 35–54, 55–74, >75 years) to allow comparison across studies and population statistics. Death rates were based on overall deaths by the end of the 7-year follow-up and also calculated for each year after injury and standardized to give an annual rate per 1000 population. Death rates for Glasgow and Scotland were taken from tables published by the GROS for 2002 (the end point of the 7-year follow-up).
|
Results |
|---|
|
TopSummaryIntroductionMethodResultsDiscussionReferences |
|---|
At the end of year 7, 206/767 (27%) had died. Of these 73 (9.5% of the sample; 35.4% of all deaths) died in the first year after injury. On average 22 died in each subsequent year (3.2% of sample per year, 10.6% of all deaths) with no trend obvious between years 2 and 7 (Table 1).
|
Within year 1 (Fig. 1), more than half of the 73 deaths were in the first (n = 31) or second (n = 12) month after injury; for months 3–12 the rate per month was much lower (3.0 per month; range 1–6) and was lower again in months 13–84 (1.8 per month; range 0–7).
|
Characteristics of early and later deaths compared to survivors (Table 2)
A head injury was recorded as the primary cause of death in 23 of the original sample of 767; of these 18 died in months 1–2, the rest dying 8, 12, 13, 34 and 37 months after injury. Few differences were found between those who had died and the survivors in terms of characteristics at injury between months 1–2 and 3–12 (P > 0.05). An exception was severity of head injury (Chi2 = 12.14, df = 2, P = 0.002). In months 1–2, 19% (19/100) of severe head injuries died, compared to 9% (12/133) of moderate and 2% (10/507) of mild. In months 3–12 there was little difference (6%, 5/81 severe; 4%, 5/121 moderate and 4%, 19/497 mild).
|
The characteristics associated with those who had died or had survived at the end of year 1 were similar, except that those dying within year 1 more often had a severe head injury (Chi2 = 34.68, df = 2, P < 0.001) and less social deprivation (Mann–Whitney, z = 2.19, P = 0.025) than survivors. At the end of year 7 more of those who died than survived, had lived alone (Chi2 = 25.31, df = 1, P < 0.001) and habitually drank alcohol to excess (Chi2 = 11.66, df = 1, P < 0.001). Some characteristics that distinguished those who had died from those still alive were the same at both time points (i.e. 1 and 7 years). Those who died were older at injury (Mann–Whitney, z = 13.71, P < 0.001) and pre-injury, more often had a history of admission for brain illness (Chi2 = 16.19, df = 1, P = 0.001) or physical limitation (Chi2 = 84.84, df = 1, P < 0.001), injury was more often the result of a fall and less often an assault (Chi2 = 52.19, df = 3, P < 0.001).
Disability was assessed in survivors 1 year after injury using the Glasgow Outcome Scale-Extended (Thornhill et al., 2000). Severe disability at 1 year was more often associated with death (41%; 36/89) than survival (21%; 79/379) by 7 years (Chi2 = 15.03, df = 2, P = 0.001). This association between severe disability at 1 year and death by 7 years remained after excluding people with pre-injury physical limitations (40%, 11/29 dead; 19%, 47/250, survived. Chi2 = 5.98, df = 2, P = 0.050), or pre-injury hospital admission for other brain injury (41%, 19/46 dead; 21%, 58/271 survived. Chi2 = 8.51, df = 2, P = 0.014).
For those who survived year 1, there was also information about whether rehabilitation was received or not since injury and about employment status. In survivors at 1 year, the frequency of survival or death at 7 years was not associated with receiving rehabilitation or not within year 1 (Chi2 = 0.00, df = 1, P = 0.989). However, it is of note that rehabilitation available at that time was often provided by single disciplines and rarely involved a co-ordinated multidisciplinary inpatient or community programme. In terms of employment status in those surviving at 1 year, those who had died by 7 years had more often retired and were less often in full-employment, both pre-injury (Chi2 = 78.294, df = 6, P < 0.001) and 1 year after injury (Chi2 = 77.503, df = 6, P < 0.001).
Death rates in Scotland, in Glasgow and in the sample (Fig. 2)
Throughout the 7-year follow-up period, people with head injury died at a higher rate (9.5% in year 1; 3.2% thereafter) than for the general population aged over 14 in Scotland (1.4%) or in the Greater Glasgow area (1.5%). A higher proportion of the sample aged over 54 years (56%) than those aged 15–54 years at injury (14%) had died by year 7. The death rate was particularly high in months 1–2 in those aged over 54 years (Table 3). However, relative to the death rates for Scotland or Glasgow, the death rate in the sample is higher in those aged 15–54 years than in those over 54 years; this is most marked in the first 2 months after injury, but continues in months 3–12 and years 2–7. For example, if aged 15–54 years and having sustained a head injury 13–84 months before, the risk of dying is 7 times higher amongst survivors of a head injury than for the general population in this age group in Glasgow.
|
|
For the older group (>54 years) the characteristics of the dead and survivors were similar, except that pre-injury physical limitations were more common in those who died (Chi2 = 9.49, df = 1, P = 0.002). Those who died in the younger group (15–54 years) differed from survivors in several ways, and more often had a severe head injury (Chi2 = 6.01, df = 1, P = 0.049), an injury caused by a fall (Chi2 = 12.23, df = 3, P = 0.007), a pre-injury history of physical limitations (Chi2 = 27.91, df = 1, P < 0.001) or pre-injury health problems (mental problems requiring hospital treatment, Chi2 = 9.29, df = 3, P = 0.026; head injury requiring neurosurgery, Chi2 = 25.25, df = 3, P < 0.001; admission for a brain illness, Chi2 = 28.50, df = 3, P < 0.001) and lifestyle factors (habitual excess alcohol use, Chi2 = 25.85, df = 1, P < 0.001; lived alone, Chi2 = 7.63, df = 1, P = 0.006). Even after exclusion of people with a significant pre-injury health history (physical limitations, mental problems requiring hospital treatment, head injury requiring neurosurgery or admission for brain illness), the risk of death in those aged under 55 years at injury remains higher than for the Glasgow population (odds ratio 3.11; 95% CI 1.49–5.71), whereas the increase in risk is marginal for those aged over 54 years (odds ratio 1.28; 95% CI 0.75–2.05).
After accounting for death rates by gender in the Glasgow population, the risk of death 7 years after head injury remains slightly higher for females than males (odds ratio 1.66; 95% CI 1.08–1.86).
Cause of death
The 92.3% of deaths fell within six ICDH categories: neoplasms (11.7%); mental/behavioural disorders (e.g. alcohol or drugs or dementia; 7.8%); diseases of circulatory system (27.7%); diseases of respiratory system (16.5%); diseases of digestive system (8.7%) and external causes (19.9%). These are the same mechanisms as the most common causes of death in the general population in Scotland, but with higher frequencies (Table 4).
|
The especially high death rate from external causes is in part explained by early deaths in the first 2 months after injury (n = 25) ‘External Causes’ includes intentional self-harm and ‘event of undetermined intent’. Both of the people in the study who died from self-harm were young males who had a mild head injury and died more than 4 years later; one had a history of hospital admission for mental health problems. The difference between the actual and expected (Glasgow) deaths from self harm was not significant (OR 2.87; 95% CI 0.35–10.35). There were nine cases of undetermined intent (3-injury by unspecified means, 1-injury by jumping from a height, 2-fall from a height, 1-opiate poisoning, 2 with no cause discovered). It is possible that in some of these instances there could have been intentional self-harm. Four of these deaths were directly attributable to the head injury. Hence, the maximum number who may have died from intentional self-harm following hospital discharge after the head injury is seven (two from known self-harm plus five undetermined intent) and if so, this would be more than expected for Glasgow (OR 6.86; 95% CI 2.75–14.14). Drug-related deaths in the study (n = 8, mainly opiates) seem more frequent than expected for Glasgow (OR 9.93; 95% CI 4.29–19.58). Large confidence intervals for possible deaths from self-harm and drugs make these findings tentative.
In those surviving more than 2 months and dying within 7 years, cause of injury (fall, assault, RTA or other) was not randomly distributed across the six main cause of death categories (Chi2 = 35.04, df = 18, P < 0.009). Of 108 people with a head injury caused by a fall, none died as a result of a subsequent fall and death resulted from circulatory problems in 48 (40%). Of those injured in an assault, 11/24 (46%) died from mental/behavioural or external causes, 10/20 (50%) injured by ‘other’ causes died from respiratory or digestive disease and the 10 people injured in a RTA were more evenly distributed across the cause of death categories. Social deprivation (see Whitnall et al., 2006) was divided into ‘higher’ and ‘lower’ using a median split. The main categories of cause of death in Table 4 were not associated with ‘higher’ or ‘lower’ social deprivation for year 1 or years 2–7 (Kruskal–Wallis, Chi2 = 2.278, df = 6, P = 0.892).
Head injury, seizures and dementia
Epileptic seizures were the cause of death for three people (time injury–death 9, 11 and 37 months). Forty-five people were recorded as suffering from generalized seizures during hospital admission after the head injury; of these, 15 died (five in year 1). Generalized seizures were not more commonly associated with death than survival in year 1 (Chi2 = 0.419, df = 1, P = 0.518) or 13 m–7 years (Chi2 = 0.926, df = 1, P = 0.326). Dementia was a cause of death in three people (age at death 76, 81 and 96 years and time between death and injury 80, 2 and 51 months, respectively).
|
Discussion |
|---|
|
TopSummaryIntroductionMethodResultsDiscussionReferences |
|---|
Main findings
People admitted to hospital in Glasgow after a head injury faced a risk of death over the next 7 years that was several times higher than for the general population in Scotland or Glasgow. This unambiguous finding was not solely explained by death related to direct effects of the head injury during hospital admission nor was it related consistently over time to the severity of the head injury, the age of the victim or the extent of subsequent sequelae. The death rate was highest in year 1, especially in months 1–2 when it was associated with severe head injury. There was no trend in death rate in years 2–7, when death was associated with lifestyle (living alone and habitual alcohol abuse) and being disabled 1 year after injury and not with severity of head injury.
Those who died within 7 years of head injury were often older at injury or death than survivors. However, the risk of death was higher relative to the Glasgow or Scottish populations in those aged <55 years at the time of the head injury. Hence, for every 10 people aged 15–54 years in the general population in Glasgow who died each year, there were 85 people in this age group who died after head injury, whereas for those aged over 54 years, 19 died after head injury.
Relation to other studies
Consistent with findings in the present study, mortality after head injury has been previously reported to be high after hospital discharge (Achte et al., 1971; Walker et al., 1971; Lewin et al., 1979; Weiss et al., 1982; Rish et al., 1983; Conroy and Kraus, 1988; Vollmer et al., 1991; Kilaru, 1996; Masson et al., 1997; Strauss et al., 1998; Signorini et al., 1999; Kraus et al., 2000; Baguley et al., 2000; Shavelle et al., 2001; Teasdale and Engberg, 2001; Hukkelhoven et al., 2003; Pentland et al., 2004; Brown et al., 2004; Harrison-Felix et al., 2004). There is danger of inherent sampling bias and limitations in the generality of findings when studies are not prospective, use unrepresentative samples from clinic or rehabilitation populations, military casualties or specific databases or if the sample is restricted by age or severity of brain injury. A further important consideration is the comparison between actual deaths after head injury and deaths expected in the general population from which the head injuries come. The present study is unique in taking all of the above factors into account.
Cardiovascular disease is a frequent cause of death in follow-up studies on head injury (Baguley et al., 2000; Shavelle et al., 2001; Pentland et al., 2004) and was a frequent cause of death in our study (the most frequent cause in years 2–7). Perhaps associated with this, reduced mobility and self-care after head injury are emphasized as predictors of mortality after head injury (Strauss et al., 1998; Baguley et al., 2000; Shavelle et al., 2001). Other studies have found that health, lifestyle and disability pre-injury and post-injury are linked to death or survival (Vollmer et al., 1991; Teasdale and Engberg, 2001). Our findings accord with this, but also indicate that risk of death remains elevated in people without these characteristics pre-injury. Greater social deprivation is also associated with a higher risk of death in the general population (McLoone and Body, 1994), and from head injury during hospital admission (Dunn et al., 2003), but in our study, was not associated with death after head injury and or with cause of death.
Older age at injury has often been associated with an especially high risk of death after head injury (Susman et al., 2002) including during hospital admission (Mosenthal et al., 2002), and it has been suggested that this may be due to an altered patho-physiological response in the ageing central nervous system (Vollmer et al., 1991). Reports of a higher risk of death in older adults compare rates within their head injury population (Masson et al., 1997; Signorini et al., 1999; Susman et al., 2002; Mosenthal et al., 2002; Hukkelhoven et al., 2003). Death was associated with older age at injury in our study, however, a striking new finding came from comparing the death rates observed in this cohort to those expected for similarly aged groups in the Scottish or Glasgow population—whereby risk of death was much greater in people with head injury if aged under 55 years than older. Few other studies have taken account of population statistics. Of those who have, two found no elevated risk of death more than 6 m after head injury (Conroy and Kraus, 1988; Brown et al., 2004). A third found mortality increased by 1.95 times in those surviving to 1 year after injury, with older people being at greater risk (Harrison-Felix et al., 2004); this overall rate is similar to our own finding (2.15 times), but in younger adults we found the relative risk to be much higher (7.12 times). Also of note is that a higher risk of death is present even in people aged 15–54 years with no significant pre-injury health history. This could suggest that the head injury itself played a causal role, for example by provoking a change in lifestyle or that it identifies a group in the population at higher risk of death for other reasons. This is an important finding because of the possibility of reducing this risk, and further research is required.
Some report that females are at risk of poorer outcome (death or disability) than males (Farace and Alves, 2000) particularly in the first year after injury (Kraus et al., 2000), but others do not find this (Signorini et al., 1999; Baguley et al., 2000; Ritchie et al., 2000; Coimbra et al., 2003; Brown et al., 2004; Mushkudiani et al., 2007). Some suggest that outcome is better in females (Gujral et al., 2006) and hypothesize that progesterone is neuroprotective (Groswasser et al., 1998). There was no evidence to support this in the present study where the risk of death was slightly higher for females.
A higher incidence of death from suicide in people with head injury has been reported (Tate et al., 1997; Baguley et al., 2000; Teasdale and Engberg, 2001; Pentland et al., 2004). We found no clear evidence to support the contention, as have others (Lewin et al., 1979; Roberts, 1979; Shavelle et al., 2001; Harrison-Felix et al., 2006). However, the low frequency of suicide and the difficulty in assigning cause of death (or intent more specifically) in some cases makes a definite conclusion inadvisable, as is also the case for death caused by seizures (Lewin et al., 1979; Roberts, 1979; Shavelle et al., 2001; Harrison-Felix et al., 2006) or dementia (Fleminger, 2002). In addition, the relationship between the effects of head injury and suicide is difficult to disentangle from other socio-demographic risk factors (such as young adult, male, abusing alcohol or drugs) for both head injury and suicide (Teasdale and Engberg, 2001). Finally it is of note that the occurrence of suicide in one long-term study did not peak until around 15–19 years after injury (Achte et al., 1971). All of this indicates a need for further, systematic long-term outcome research.
Strengths/limitations
The design of the Hospitalised Head Injury Study has two key advantages. First, the prospective identification of people and collection of clinical information at the time of admission to hospital following head injury. This minimized bias in characterization and subsequent stratification. Second, every effort was made to identify all relevant patients. This produced a sample that was substantially more complete and accurate than if routine records had been used and, importantly was also comprehensive in covering all severities of injury. The mortality rates reported here are those for the sample that was selected for follow up at 1 year post-injury and not for the original cohort of almost 3000. The structured follow-up sample included all of those with severe head injury, and mild head injury was proportionately under-represented, but this is not an issue for long-term follow-up given that initial severity of injury was not associated with death or survival after the first year. Inclusion of mild head injury in the sample makes its findings more representative of the wider picture. The method of identifying deceased patients may be subject to inaccuracies and relies on the GROS being informed of all deaths at the time of the matching procedure. For example, the GROS returned 58 ‘No Matches’ and it is not known if any of these people moved away and then died in another part of the UK, or overseas. The design of the study limits what can be concluded about the impact of head injury on life expectancy. It demonstrates that people who sustain a head injury are at greater risk of death than the general population from which they come, but the extent to which the head injury is a marker for risk-related, past life and social factors or itself provokes changes that increase the risk of death is yet to be determined. Future studies might consider a prospective design that incorporates an age-matched control group of hospital admissions with no head injury, although reason for admission is likely to be a key factor as is previous medical history and in practice the creation of a control of this kind may be difficult to achieve.
Implications
The clear definition that, even after surviving an acute head injury victims face a substantially increased mortality for at least 7 years, especially if younger, points to the need to understand reasons. Our findings suggest that, interventions designed to effect social and lifestyle changes, including general health preventative measures could have considerable health benefit, and should be aimed at people who survive after admission to hospital with a head injury.
|
Acknowledgements |
|---|
Thanks are due to Elaine Mulligan, Sharon Thornhill and Leigh Whitnall-Pate for early work on this project and to Paul McKenzie for administrative assistance. Dr A. Maas and Professor G. Murray provided helpful comments on a draft of the manuscript. The Chief Scientist Office for Scotland (CZH/4/48) funded the study.
|
References |
|---|
|
TopSummaryIntroductionMethodResultsDiscussionReferences |
|---|
Achte KA, Lonquist J, Hillbom E. Suicide following war brain injuries. Acta Psychiatrica Scandanavica, Suppl (1971) 225:1–94.
Baguley I, Slewa-Younan S, Lazarus R, Green A. Long term mortality trends in patients with traumatic brain injury. Brain Inj (2000) 14:505–12.[CrossRef][Web of Science][Medline]
Brown AW, Leibson CL, Malec JF, Perkins PK, Diehl NN, Larson DR. Long term survival after traumatic brain injury: a population based analysis. Neurorehabilitation (2004) 19:37–43.[Web of Science][Medline]
Coimbra R, Hoyt DB, Potenza BM, Fortlage D, Hollingsworth-Fridlund P. Does sexual dimorphism influence outcome of traumatic brain injury patients? The answer is no! J Trauma (2003) 54:689–700.[Web of Science][Medline]
Conroy C, Kraus JF. Survival after brain injury. Neuroepidemiology (1988) 7:13–22.[CrossRef][Web of Science][Medline]
Dunn L, Henry J, Beard D. Social depreivation and adult head injury: a national study. J Neurol Neurosurg Psychiatry (2003) 74:1060–4.
Farace E, Alves WM. Do women fare worse: a meta-analysis of gender differences in traumatic brain injury outcome. J Neurosurgery (2000) 93:539–45.[Web of Science][Medline]
Fleminger S, Oliver DL, Lovestone S, Rabe-Hesketh S, Giora A. Head injury as a risk factor for Alzheimer's disease. J Neurology Neurosurgery Psychiatry (2002) 74:857–62.
Gennarelli TA, Champion HR, Sacco WJ, Copes WS, Alves WM. Mortality of patients with head injury and extracranial injury treated in trauma centres. J Trauma-Inj Infect Crit Care (1989) 29:1193–201.[CrossRef]
Groswasser Z, Cohen M, Keren O. Female TBI patients recover better than males. Brain Inj (1998) 12:805–8.[CrossRef][Web of Science][Medline]
Gujral IB, Stallones L, Gabella BA, Keefe TJ, Chen P. Sex differences in mortality after traumatic brain injury, Colorado 1994–1998. Brain Inj (2006) 20:283–91.[CrossRef][Web of Science][Medline]
Harrison-Felix C, Whiteneck G, DeVivo MJ, Hammond FM, Amitabh J, et al. Mortality following rehabilitation in the traumatic brain injury model systems of care. Neurorehabilitation (2004) 19:45–54.[Web of Science][Medline]
Harrison-Felix C, Whiteneck G, DeVivo MJ, Hammond FM, Amitabh J. Causes of death following 1 year postinjury among individuals with traumatic brain injury. J Head Trauma Rehabil (2006) 21:22–33.[Web of Science][Medline]
Hukkelhoven CW, Steyerberg EW, Rampen AJ, et al. Patient age and outcome following severe traumatic brain injury. J Neurosurg (2003) 99:666–73.[Web of Science][Medline]
Jennett B. Epidemiology of head injury. J Neurol Neurosurg Psychiatry (1996) 60:362–9.
Kilaru S, Garb J, Emhoff T, Fiallo V, Simon B, Swiencicki T, et al. Long-term functional status and mortality of elderly patients with severe closed head injuries. J Trauma (1996) 41:957–63.[Web of Science][Medline]
Kraus J, Peek-Asa C, McArthur D. The independent effect of gender on outcomes following traumatic brain injury: a preliminary investigation. Neurosurg Focus (2000) 8. Article 5.
Lewin W, Marshall TFC, Roberts AH. Long-term outcome after severe head injury. BMJ (1979) 2:1533–8.
Masson F, Vecsey J, Salmi LR, Dartigues JF, Erny PH, Maurette P. Disability and handicap 5 years after a head injury: a population based study. J Clin Epidemiol (1997) 50:595–601.[CrossRef][Web of Science][Medline]
McLoone P, Body FA. Deprivation and mortality in Scotland 1981 and 1991. BMJ (1994) 309:1465–70.
Mosenthal AC, Lavery RF, Addis M, Sanjeev K, Ross S, Marburger RBS, et al. Isolated traumatic brain injury: age is an independent predictor of mortality and early outcome. J Trauma (2002) 52:907–11.[Web of Science][Medline]
Mushkudiani NA, Engel DC, Steyerberg EW, Butcher I, Lu J, Marmarou A, et al. Prognostic value of demographic characteristics in traumatic brain injury: results from the IMPACT study. J Neurotrauma (2007) 24:259–69.[CrossRef][Web of Science][Medline]
Pentland B, Hutton LS, Jones PA. Late mortality after head injury. J Neurol Neurosurg Psychiatry (2004) 76:395–400.[CrossRef][Web of Science]
Rish BL, Dillon JD, Weiss GH. Mortality following penetrating craniocerebral injuries. An analysis of the deaths in the Vietnam Head Injury Registry population. J Neurosurg (1983) 59:775–80.[CrossRef][Web of Science][Medline]
Ritchie PD, Cameron PA, Ugoni A, Kaye AH. A study of the functional outcome and mortality in elderly patients with head injuries. J Clin Neurosci (2000) 7:302–4.
Roberts AH. Severe accidental head injury: an assessment of long-term prognosis (1979) London: MacMillan Press.
Shavelle RM, Strauss D, Whyte J, Day SM, Yu YL. Long-term causes of death after traumatic brain injury. Am J Phys Med Rehabil (2001) 80(7):510–6.[CrossRef][Web of Science][Medline]
Signorini DF, Andrews PJD, Jones PA, Wardlaw JM, Miller JD. Predicting survival using simple clinical variables: a case study in traumatic brain injury. J Neurol Neurosurg Psychiatry (1999) 66:20–5.
Sosin DM, Sniezek JE, Waxweiler RJ. Trends in death associated with traumatic brain injury, 1979 through 1992: success and failure. J Am Med Assoc (1995) 273:1778–80.
Strauss DJ, Shavelle RM, Anderson TW. Long term survival of children and adolescents after traumatic brain injury. Arch Phys Med Rehabil (1998) 79:1095–100.[CrossRef][Web of Science][Medline]
Susman M, DiRusso SM, Sullivan T, Risucci D, Nealon P, Cuff S, et al. Traumatic brain injury in the elderly: increased mortality and worse functional outcome at discharge despite lower injury severity. J Trauma Inj Infec Crit Care (2002) 53:219–24.[CrossRef]
Tagliaferri F, Compagnone C, Korsic M, Servadei F, Kraus J. A systematic review of brain injury epidemiology in Europe. Acta Neurochirurgica (Wien) (2006) 148:255–68.[CrossRef]
Tate R, Simpson G, Flanagan S, Coffey M. Completed suicide after traumatic brain injury. J Head Trauma Rehabil (1997) 12:16–28.[CrossRef][Web of Science]
Teasdale TW, Engberg AW. Suicide after traumatic brain injury, a population study. J Neurol Neurosurg Psychiatry (2001) 71:436–40.
Thornhill S, Teasdale GM, Murray GD, McEwen J, Roy CW, Penny NI. Disability in young people and adults one year after head injury: prospective cohort study. BMJ (2000) 320:1631–5.
Vollmer DG, Torner JC, Eisenberg HM, Foulkes MA, Marmarou A, Marshall LF. Age and outcome following traumatic coma: Why do older patients fare worse? J Neurosurg (1991) 75:S37–49.[Web of Science]
Walker AE, Leuchs HK, Lechtape-Gruter H, et al. Life expectancy of head injured men with and without epilepsy. Arch Neurol (1971) 24:95–100.
Weiss GH, Caveness WF, et al. Life expectancy and causes of death in a group of head injured veterans of world war I. Arch Neurol (1982) 39:741–43.
Whitnall L, McMillan TM, Murray G, Teasdale G. Disability in young people with head injury: a 5–7 year follow-up of a prospective cohort study. J Neurol Neurosurg Psychiatry (2006) 77:640–5.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. J. Hilz and S. Schwab Stroke-Induced Sudden-Autonomic Death: Areas of Fatality Beyond the Insula Stroke, September 1, 2008; 39(9): 2421 - 2422. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||