Incidence

According to the most recent figures from the CDC, the average estimated incidence of TBI in the United States between the years 2002 and 2006 was 1,691,481 (576.8 per 100,000). Of these 1,364,797 (465.4 per 100,000) were treated in the emergency department (ED) and released; 275,146 (93.8 per 100,000) were hospitalized and discharged alive; and 51,538 (17.6 per 100,000) died.¹² Earlier examination of CDC data between 1995 and 2001 estimated a total of 1,396,000 (506.4 per 100,000) TBIs each year with 1,111,000 (403.1 per 100,000) ED visits, 235,000 (85.2 per 100,000) hospitalizations, and 49,900 (18.1 per 100,000) deaths.¹⁴ These data from the CDC do not include individuals that were treated at outpatient facilities or who did not seek treatment.

Previous reports from the United States that used the 1991 National Health Interview survey of 46,761 households estimated there were approximately 1.5 million (618 per 100,000) individuals who sustained nonfatal brain injuries in the United States and of those approximately 25% did not seek any type of treatment.¹⁵ Comparison between the Sosin and colleagues report¹⁵ and the reports from the CDC data are difficult to make based on methodological differences. However, comparison between the CDC reports from Langlois and colleagues¹⁴ to that of Faul and colleagues¹² clearly shows that the average annual estimate of ED visits and hospitalizations has increased but the rate of TBI related deaths has decreased. This was investigated more closely in a report of death rates from 1997 to 2007 that showed an 8.2% decrease, from 19.3 to 17.8 per 100,000, during this time and the authors noted that this follows a long-standing trend of decreases in TBI-related deaths over time.¹⁶

A review of European epidemiological studies by Tagliaferri and colleagues found a wide range in the incidence in TBI.¹⁷ The authors reviewed 23 epidemiological reports from various European countries between 1980 and 2003. Methodology varied considerably between the studies, as did the estimated rates of TBI. One study from a province of Spain estimated an annual average to be 91 per 100,000, whereas a study from Sweden estimated 546 per 100,000. The authors removed the extreme reports from the study and found an aggregate of 235 per 100,000 TBIs annually.

The aforementioned studies constitute some of the best epidemiological data on incidence of TBI, but the data likely grossly underestimates the incidence of mTBI.⁵ A World Health Organization (WHO) systematic review of the mTBI literature found that 70% to 90% of TBI was mild in nature and that hospital-treated mTBI was approximately 100 to 300 per 100,000 in the studies it reviewed. However, given the undertreatment and reporting of mTBI, the WHO estimated that the true yearly incidence was likely 600 per 100,000.¹⁸

A population-based study in New Zealand attempted to correct for underreporting by using both prospective and retrospective surveillance systems in an attempt to register all instances of TBI over a 1-year period between March 2010 and February 2011. The study was based in the city of Hamilton, which has a population of 129,249 and the surrounding rural area with a population of 43,956. The investigators reportedly enrolled all healthcare providers (e.g., family physicians, health centers, hospitals, ambulance services), the local prison, community services (e.g., school, sports clubs), and used national healthcare databases and the death registry to assist in identifying cases. Regular contact was kept among these sources throughout the year and each identified case was crosschecked to ensure there were no duplicates. During this time, 1369 TBIs were identified, which the authors concluded to be an annual incidence rate of 790 per 100,000.¹⁹ This rate is considerably higher than most epidemiological studies and the authors asserted that this was due to a larger number of mild traumatic brain injuries captured that typically go unaccounted for in most other epidemiological studies.

Severity

Data regarding the severity of TBI is difficult to ascertain, given that classification systems are not consistent across studies or sometimes even within studies. It is commonly accepted that the majority of TBIs are mild in severity and make up between 70% and 90% of all TBIs.²⁰ This number is likely an underestimation, given that many mTBIs are thought to go untreated and therefore unreported. In Feigin and colleagues,¹⁹ mTBI made up nearly 95% of the total sample. Another confounding factor when considering injury severity is that individuals who sustained more significant brain injury or other trauma are commonly sedated, which complicates injury classification. In one study of adults receiving inpatient rehabilitation services, 19.2% of the sample was classified as severe, 10.3% was moderate, and 16.4% of the sample was chemically sedated. The injury severity and reason for sedation in those individuals is not entirely clear, although the authors indicated that most were thought to be in the severe range.²¹

In addition to underestimating the number of mild injuries, the current epidemiological literature likely overestimates the number of severe injuries as many studies utilize hospital admission records to classify injury. Similar to Cuthbert and colleagues,²¹ studies that utilize hospitalization records report that 20% of injuries are classified as severe²² while other authors reported that 80% of hospitalized individuals sustained mild injuries, 10% sustained moderate injuries, and 10% were severe. These figures are considerably higher than the report by Feigin and colleagues,²¹ and also well above a review of European epidemiological studies that calculated a TBI severity ratio of 22:1.5:1 for mild, moderate, and severe injuries, respectively.

Risk Factors and Characteristics

Alcohol

The positive association between blood alcohol concentration (BAC) and many types of injuries is well established. In terms of brain injury, it has been shown that a high percentage of individuals seek medical attention following a TBI have a positive BAC, with figures ranging from 56% to 72% in various United States samples.²⁶ In addition, a large number of those individuals (36% to 55%) were legally intoxicated at time of injury.^27–30 European epidemiological studies also show a strong relationship between alcohol consumption and TBI, though the figures are less consistently high as the data from U.S. populations, with reports ranging from 24% to 51% who were positive for alcohol.¹⁷ Langlois and colleagues reported that alcohol use was reported in 21% of motor vehicle occupants hospitalized with a TBI, and over 12% had a BAC above the legal limit.³¹

Recurrent TBI

Recurrent TBI, especially mTBI, has become a topic of considerable interest in recent years due to concern that TBI increases the risk of cognitive impairment later in life³² or possible neurodegenerative conditions.³³ Though skepticism exists regarding the link between mTBI and the development of neurodegenerative conditions³⁴ it has been shown that recurrent mTBI is associated with prolonged recovery and that athletes who have a history of mTBI may be at greater risk of sustaining future injuries.³⁵ Annegers and colleagues conducted the first population based study of recurrent TBI and found that the relative risk of a second TBI among those with an earlier TBI was 2.8 to 3 times greater than the noninjured sample.³⁶ Additionally, in those that sustained a second head injury the risk of sustaining a third head injury was 7.8 to 9.3 times that of an initial head injury in the population. Recurrent TBI has been found to be related to alcohol abuse in a number of studies²² with one study indicating that if an individual’s first injury after age 12 was alcohol related they were at a fourfold greater risk of repeat head injury by age 34. Recurrent TBI has long been a concern in sports with a recent epidemiological study of sport concussion in high school athletes in the United States showing that 13.2% of all concussions were recurrent, and consistent with previous findings the symptoms of recurrent concussion took longer to resolve.³⁷

External Causes

According to the most recent data from the CDC, falls are the most common cause of TBI, with an estimated annual average of 523,043. The rate of fall-related TBI was greatest among children aged 0 to 4 years (839 per 100,000) and adults age 75 and older (599 per 100,000). Falls accounted for approximately half of the TBIs in children age 0 to 14 and approximately 60% of adults aged 65 and older. The second most common cause is motor vehicle traffic accidents (estimated annual average 292,202), followed by struck by/against events (estimated annual average 169,625) and assault (estimated annual average 169,625).¹² Motor vehicle accidents resulted in the largest percentage of TBI-related deaths at 31.8%.

Few European epidemiological studies have investigated external causes, but most reports indicate that motor vehicle accidents are the most common events leading to TBI, followed by falls.¹⁷ It should be noted that the external cause of TBI varies considerably by country due to a myriad of factors, including economic status. Lower and middle-income countries that had not had access to motor vehicles are showing sharp increases in motor-vehicle-related injuries and deaths, while wealthier countries are developing better safety standards and road laws that have led to a decrease in motor-vehicle-related mortality over the years.³⁸ These factors need to be better understood to understand the epidemiological impact of TBI around the world.

TBIs related to engagement in sports and recreational activity are a major cause of TBI, but are likely underreported in the literature given that most of the injuries are mild in severity and untreated. Previous epidemiological studies estimated approximately 300,000 sports-and recreation-related TBIs per year.³⁹ However, it has been noted that Thurman and colleagues³⁹ only included TBIs for which the person reported a loss of consciousness, which is considered a very small minority (10%) of TBIs related to sports and recreation.^40,41 Langlois and colleagues indicated that if all injuries are taken into account, including those who do not seek medical attention, a more accurate estimate of 1.6 to 3.8 million sports-and recreation-related concussions occur annually.² The data on sports-related concussion is likely to improve in the coming years as identification of concussion by athletic trainers, coaches, athletes, and medical personnel improves and more athletes seek treatment. A recent study of emergency departments found that sports-related concussion visits doubled for children aged 8 to 13 and increased by more than 200% in individuals aged 14 to 19 from 1997 to 2007.⁴²

Abusive head trauma is the leading cause of serious head injury and death in children age 2 years and younger. Keenan and colleagues prospectively collected data from nine hospital pediatric intensive care units from January 1, 2000, to December 31, 2001.⁴³ Results showed that the incidence of inflicted TBI in the first 2 years of life was 17 per 100,000 person-years, with infants having a higher incidence (29.7 per 100,000) relative to children in the second year of life (3.8 per 100,000). To determine whether an injury was inflicted the authors relied on confession from the parents or medical/social service determination of abuse so it is likely that these are underestimates of the true incidence. A more recent study using a national sample of inflicted injury in children found a slightly higher rate of 18.7 per 100,000 for children less than 2 years of age.⁴⁴

TBI associated with violence against adults is also underreported and underidentified due to many factors, including low reporting rates in domestic violence and other factors associated with the types of trauma sustained during assaults.²² Previous data from the CDC estimated that at least 156,000 TBI-related emergency department visits, hospitalizations, and deaths each year are associated with all types of assaults² with more recent estimates showing approximately 169,625 TBIs per year. According to these data, assaults account for approximately 10% of total TBIs.¹² A study based on the New Zealand population found that assaults comprised almost 17% of all TBIs with an estimated annual incidence of 132 per 100,000,¹⁹ which is considerably higher than estimates from the United States. According to the CDC data, the highest risk of sustaining a TBI secondary to assault is in individuals age 15 to 44, as this age group comprises 75% of assault-related head injuries.¹²

Lifetime Prevalence

The lifetime prevalence of TBI is less well studied in the epidemiological literature due primarily to the methodology employed by most studies. Lifetime prevalence of TBI refers to the number of individuals who have ever experienced a TBI. In a large-scale study, Winqvist and colleagues used a birth cohort design of individuals born in two provinces of North Finland in 1966.⁴⁵ The entire cohort included 12,058 residents who were followed through age 34. Results showed that the average annual incidence of TBI was 118 per 100,000 and by age 35, 3.8% of the cohort had experienced at least one hospitalization for TBI.⁴⁵ The study only included hospital and health center discharge records with outpatient services and emergency department visits of less than 24 hours being excluded from analysis, so these data are only capturing more severe injuries. A birth cohort study in New Zealand that used any type of medical attendance where TBI was diagnosed and followed through age 25 found a much higher average annual incidence rate of 1750 per 100,000. By age 25, 31.6% of the cohort had received a TBI for which they received medical care.⁴⁶ In a study by McKinlay and colleagues, only 32.9% of the individuals who sustained a TBI were admitted to the hospital for further observation or treat-ment.⁴⁶ Retrospective self-report studies by Anstey and colleagues⁴⁷ and Silver and colleagues⁴⁸ that looked at TBI with loss of consciousness indicated a lifetime prevalence rate of 5.7% to 8.5%, respectively.

Consequences of TBI

Due in large part to advances in medicine and safety, more people are surviving TBI than ever before. Estimates of individuals in the United States who are living with long-term or lifelong disability associated with a TBI range from 3.2 million to 5.3 million.^2,49 These estimates only include individuals who were hospitalized as a result of their injuries and therefore are biased toward more severe injuries. The Glasgow Outcome Scale (GOS) is an unsophisticated but widely used indicator of disability or residual effects used by hospitals at the time of discharge. The major classifications of the GOS are death, persistent vegetative state (i.e., patient exhibits no obvious cortical function), severe disability (i.e., patient is conscious but requires daily support due to disability), moderate disability (i.e., disabled but independent in regard to daily living needs), and good recovery (i.e., resumption of normal activities) though minor neurologic and psychological deficits may be present.⁵⁰ Data from a 14-state CDC-funded TBI surveillance study found that 17% of hospital discharges were reported as having moderate to severe disability based on the GOS.³¹ Krause and colleagues estimated that the total number of disabilities from new brain injuries for the year 2000 was 98,560 or a rate of approximately 35 per 100,000 of the population.²⁶ A population-based sample of persons with TBI from the South Carolina Traumatic Brain Injury Follow-up Registry was used to develop a predictive model of TBI that estimated 43.3% of hospitalized TBI survivors in 2003 experienced TBI-related long-term disability, which in turn estimated that 124,626 Americans per year who likely need rehabilitative or supportive services following a TBI.⁵¹

Disability associated with mTBI is difficult to calculate given that only about 10% are hospitalized and therefore not considered in many epidemiological studies. However it has been well established that mTBI results in physical, cognitive, psychological, and social dysfunction that leads to lost productivity and decreased quality of life (for a review, see McCrea⁶). The impact on occupation has been reported to be comparable to general trauma patients relative to the duration of time off work and reported problems upon returning to work.⁵² Boake and colleagues showed that most mTBI patients treated and released from emergency departments did not return to work until 1 to 3 months after the injury and there was virtually no difference between those who were treated and released from the emergency department (46% working at 1 month, 66% at 3 months, and 68% at 6 months) versus those who were hospitalized (39% working at 1 month, 62% at 3 months, and 71% at 6 months).⁵²

Economic Cost

The real economic cost of TBI is difficult to calculate given the number of untreated TBIs that may lead to missed workdays or decreased productivity at work. Additionally, it is difficult to truly appreciate and quantify the impact that TBI has on caregivers and family members. Langlois and colleagues²² extrapolated data from Finkelstein and colleagues⁵³ to calculate the estimated economic burden of TBI in 2009 dollars. The authors estimated that the total costs of fatal, hospitalized, and nonhospitalized TBI in the United States totaled more than $221 billion, including $14.6 billion for medical costs, $69.2 billion for work lost costs, and $137 billion for the value of lost quality of life.²² A report by the CDC indicated that the economic burden of TBI in the United States was estimated to be $56 billion, with mTBI accounting for $16.7 billion of that figure.⁵⁴

Incidence

According to the most recent report by the DoD, there have been a total of 294,172 TBIs from 2000 through the fourth quarter of 2013.¹³ The most common injury is concussion/mTBI, which accounted for 242,676 (82.5%) of the injuries, followed by moderate TBIs (23,754; 8.1%), not classifiable (20,433; 6.9%), penetrating (4389; 1.5%), and severe (2920; 1%). When the individual years are considered, the total prewar incidence of TBI was 10,958 for 2000. The years following the commencement of OEF were not considerably different, with subsequent increases of about 3% to 9% each year until 2006, when 17,037 injuries were reported. The number of injuries increased drastically in 2007, when 23,217 injuries were reported, which was an increase of approximately 36% from 2006. Another considerable increase of 22% was recorded the following year (28,877 injuries in 2008). TBI rates continued to increase until a peak in 2011 of 32,625 injuries, and then there was a slight decline in 2012 (the last full year of available data), when 30,406 TBIs were reported.

The distribution of injury type has not drastically changed since the start of OEF, with the vast majority (71% to 86%) of injuries being concussions/mTBI. The increase in TBI since the start of OEF has been driven almost exclusively by large increases in the concussion/mTBI group. According to prewar data in 2000, there were 275 penetrating injuries, 178 severe, 1616 moderate, and 7179 mild TBIs. In comparison, there were 324 penetrating, 261 severe, 1857 moderate, and 27,535 mild TBIs recorded in the year 2011, which had the highest rate of total TBIs. When the year 2011 and 2000 are compared, substantial increases are noted across injury types, including penetrating, severe, and moderate TBI, with increases of approximately 17%, 47%, and 15%, respectively. However, the most substantial increase, and what accounts for the large increase of total TBIs, was concussion/mild TBI injury that showed a 283% increase 2000 to 2011. This increase is probably at least partially accounted for by better identification and diagnosis of these less severe injuries. In any event, it is clear that concussion/mTBI has the largest scale impact on our service members.

External Causes

The mechanisms of injury for military personnel are generally similar to those found in civilians and include motor vehicle accidents, falls, sports/recreation related, assaults, and collisions with objects. According to a review by Langlois and colleagues,²² mechanism of injury data is typically only available for hospitalizations and is provided by the Armed Forces Health Surveillance Centers (AFHSC). Upon review of the data from the AFHSC, the authors showed that during the prewar period, land transport (~25%) and falls/miscellaneous (~24%) were the most common injury type followed by sports/recreation (9%) and nonbattle assaults (7%) among active-duty servicemen.²²

Injury data for deployed servicemen is difficult to attain, as this data is not included in the DoD reports. A study of an Army unit (3973 servicemen) following a 1-year deployment in Iraq identified 907 soldiers diagnosed with a TBI. According to survey data, TBI was most commonly caused by blast (88%) followed by motor vehicle accidents (39%), falls (20%), fragments (15.8%), and bullets (3.1%). The servicemen in the study commonly identified more than one cause of their injury, which led to the percentages being greater than 100.⁵⁸ Blast injury was clearly the most commonly reported mechanism of injury, and the high rate of blast injury is reported elsewhere.^59,60 It is now commonly held that a dual mechanism (blast and blunt) is common in the military environment.

There are several mechanisms of blast injury described in the literature, including primary, secondary, tertiary, and quaternary. Primary injuries are due to overpressurization or underpressurization shock waves caused by explosions. The shock waves move through the body from solid-and liquid-filled sections to gas-filled organs (e.g., lungs), which causes damage to the organs. Secondary injuries are due to fragments and projectiles from explosions. Tertiary injuries are caused by a displacement of air (blast wind), throwing the individual and causing them to collide with an object. Quaternary injuries are those not classified by the other three and include toxic inhalation, burns, and crush injuries.⁶¹ The exact nature of the blast injury is not typically identified, but it is clear that most are likely caused by secondary and tertiary injuries as these involve a mechanical component similar to other TBI-causing mechanisms.⁶² TBI related to primary blast injuries has been reported in the literature,⁶³ but this remains controversial given the exact mechanism of injury remains unclear.⁶⁴ Given the specific context in which blast injuries occur and the lack of reliable data regarding acute injury characteristics and mechanism of injury, these injuries are difficult to study and the epidemiological data is lacking. Future research is required, ideally employing a standard definition of injury and robust injury surveillance systems.