Accepted Manuscript

Traumatic Brain Injury Epidemiology in Brazil

Carlos Eduardo Romeu de Almeida, MD, José Lopes de Sousa Filho, MD, Jules
Carlos Dourado, MD, Pollyana Anício Magalhães Gontijo, RN, Msc, Marcos Antônio
Dellaretti, MD, PhD, Bruno Silva Costa, MD, Msc

PII: S1878-8750(15)01363-7
DOI: 10.1016/j.wneu.2015.10.020
Reference: WNEU 3315

To appear in: World Neurosurgery

Received Date: 30 May 2015

Revised Date: 30 September 2015
Accepted Date: 1 October 2015

Please cite this article as: de Almeida CER, de Sousa Filho JL, Dourado JC, Magalhães Gontijo PA,
Dellaretti MA, Costa BS, Traumatic Brain Injury Epidemiology in Brazil, World Neurosurgery (2015), doi:
10.1016/j.wneu.2015.10.020.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Traumatic Brain Injury Epidemiology in Brazil

Carlos Eduardo Romeu de Almeida, MD.*

José Lopes de Sousa Filho, MD*.

PT
Jules Carlos Dourado, MD*.

RI
Pollyana Anício Magalhães Gontijo, RN, Msc**

Marcos Antônio Dellaretti, MD, PhD***

SC
Bruno Silva Costa, MD, Msc***

U
Department of Neurosurgery, Santa Casa de Belo Horizonte,
AN
Belo Horizonte, Minas Gerais, Brazil
M

* Neurosurgery Resident at Santa Casa de Belo Horizonte

D

**Research Coordinator of the Neurosurgical Department at Santa Casa de Belo Horizonte

TE

***Professor of Neurosurgery at Santa Casa de Belo Horizonte

EP

Contact Information:
C
AC

Carlos Eduardo Romeu de Almeida (corresponding author)

E-mail: cadu_romeu@yahoo.com.br

Address: Rua Prof. Octaviano de Almeida, 131, Apto 903 - Santa Efigênia, Belo Horizonte,

MG, CEP 30260-020, Brazil.

Phone: +55 (31) 93050203

 ACCEPTED MANUSCRIPT
José Lopes de Sousa Filho

E-mail: joselopessousa@gmail.com

Phone: +55 (31) 91986518

Jules Carlos Dourado

PT
E-mail: julesdourado@gmail.com

RI
Phone: +55 (31) 91608888

SC
Pollyana Anício Magalhães Gontijo

E-mail: pollyanaam@hotmail.com

Phone: +55(31) 88746175

U
AN
M

Marcos Antônio Dellaretti

E-mail: mdellaretti@mac.com
D

Phone: +55 (31) 97663123

TE

Bruno Silva Costa

EP

E-mail: costabs@gmail.com
C

Phone: +55 (31) 87859429

AC
 ACCEPTED MANUSCRIPT
Abstract

Traumatic brain injury (TBI) stands out as a grave social and economic problem. Emerging

countries possess few epidemiological studies on the range and impact of TBI. Our study

aimed to characterize the demographic, social and economic profile of people suffering from

TBI in Brazil. Data on TBI cases in Brazil between 2008 and 2012 were collected through the

PT
website of the Information Technology Department of the Unified Health System

RI
(DATASUS) maintained by the Brazilian Ministry of Health. This database is fed by public

hospital admission authorization forms provided nationwide. There were around 125,000

SC
hospital admissions due to TBI a year, an incidence of 65.7 admissions per 100,000

inhabitants per year. Hospital mortality was 5.1/100,000/year, and the case-fatality rate was

U
AN
7.7%. The average annual cost of hospital expenses was US$ 70,960,000, with an average

cost per admission of US$568. The age group 20-29 was the most frequently admitted to
M

hospital due to TBI and presented the largest number of hospital deaths; however, the

population over 80 showed the highest admission rate, around 138/100,000/year, followed by
D

the age group 70-79. TBI should be recognized as an important public health problem in
TE

Brazil, since it is responsible for considerable social and economic costs. Besides the young

adult age group (20-29 years old), the geriatric age group is especially vulnerable to the
EP

frequent and devastating consequences of TBI. The implementation of a system of effective

C

epidemiological vigilance for neurotrauma is urgent in Brazil and other countries worldwide.
AC

Keywords: Head Trauma, Epidemiology, Prevention and Control, Economics.

 ACCEPTED MANUSCRIPT
Introduction

Traumatic brain injury (TBI) can be defined as an alteration of cerebral function or

evidence of cerebral pathology caused by an external force,(13) and is an important public

health problem worldwide. It is the principle cause of disability in children and adults under

PT
35, and seriously affects the lives of patients and their families.(5) In 2009, the Centers for

Disease Control and Prevention (CDC) estimated that there had been at least two million

RI
visits to emergency services in the USA, 300,000 hospital admissions, 52,000 deaths caused

by TBI, signifying per 100,000 inhabitants coefficients of 686 visits to emergency services,

SC
95.7 hospitalizations and 16.6 deaths due to TBI.(4) Furthermore, a meta-analysis based on

U
European studies estimated an incidence of around 235 admissions due to TBI/100,000
AN
inhabitants/year in the European Community, with a mortality rate of 15.4/100,000/year.(20)

However, emerging countries possess few epidemiological studies on the range and
M

impact of TBI. In Qatar, Bener et al. identified an incidence of 44/100,000/year from 2007 to

2009, although TBI caused by violence was excluded.(2) A study conducted in Johannesburg,
D

estimated an annual incidence of TBI in South Africa of 316/100,000 in 1991.(15)

TE

In Brazil, and in Latin American countries generally, epidemiological studies on this

EP

theme are also scarce. In 1986, in São José do Rio Preto, São Paulo, Maset et al. identified

an annual incidence of 456/100,000, with the age group 20-29 presenting an incidence of
C

710/100,000.(10) In 1991, Masini estimated the incidence of TBI in the Federal District to be
AC

341/100,000.(8) Fernandes and Silva also studied the epidemiology of TBI in Brazil based on

DATASUS data in reference to the age group 14-69 and from 2000 to 2007 and estimated an

incidence of hospital admission of around 37/100,000 inhabitants in this age bracket.(17)

Because of its high incidence, great potential for disability and impact on the

economically active population, TBI stands out as a grave social and economic problem and

is currently considered a “silent epidemic”.(18) Approximately 30% of deaths due to external

 ACCEPTED MANUSCRIPT
causes include a diagnosis of TBI, and it is estimated that around 5.3 million people in the

USA live with conditions related to this injury, including cognitive and psychological

alterations.(4,7,23)

In the Brazilian literature on the subject, no estimates of the prevalence of alterations

in health status related to TBI were found. In the USA, the cost of TBI in 2010 was estimated

PT
to be US$76.5 billion, comprising US$11.5 billion in direct medical expenses and US$64.8

RI
billion indirect expenses.(23) The majority of TBI epidemiological studies provide no

information at the population level on the presence of disability, financial impact and loss of

SC
productivity.

The World Health Organization (WHO) initially recognized the importance of

U
adopting an efficient system of epidemiological vigilance of neurotrauma in 1995, when it
AN
launched the Standards for Surveillance of Neurotrauma, defining directives for the
M

development of these systems to make it possible to measure its impact on the population and

for the development of more effective prevention methods.(22) The purpose of systems of
D

vigilance and epidemiological studies is to provide the necessary information to introduce

TE

primary (avoiding the occurrence of trauma), secondary (rapid and proper treatment to reduce

deaths and disability) and tertiary prevention (mitigating the disability and reducing the
EP

individual limitations caused by TBI).

C

Furthermore, in accordance with the WHO, to be able to properly estimate the

AC

importance of a harmful factor for public health it is necessary to describe the magnitude of

the problem, i.e. the number of cases, coefficients of incidence and mortality, the prevalence

of disability, the population exposed to greatest risk of trauma, the causes, the gravity and the

consequences (disability and costs).(22)

In terms of epidemiological vigilance and the evaluation of the impact of the

neurotrauma, the USA stands out as a model for emulation. In 1989, the Federal Interagency
 ACCEPTED MANUSCRIPT
Head Injury Task Force had already identified TBI as an important public health problem that

demanded specific epidemiological vigilance. In 1995, the CDC established directives for the

vigilance of neurotrauma nationwide, and such directives launched the basis for the Standards

for Surveillance for Neurotrauma of the WHO. In 1996, the American Congress approved a

law that, among other attributions, charged the CDC with the function of developing a system

PT
of uniform data collection for TBI cases and of conducting and financing research to identify

RI
effective strategies to reduce the incidence of the neurotrauma.(4) Since then, a vast literature

composed of reports and information bulletins has helped to clarify the dimensions and the

SC
importance of TBI, as well as to guide public policies regarding its prevention and the

reduction of its social and economic impact

U
(http://www.cdc.gov/traumaticbraininjury/factsheets_reports.html).
AN
Given the importance of the theme, the scarcity of data in the scientific literature and
M

the growing need for specific epidemiological vigilance policies for neurotrauma in Brazil

and worldwide, the objectives of this study were to evaluate the magnitude of TBI in the
D

Brazilian population by means of estimates of incidence, costs, hospitalization time, deaths,

TE

distribution by age, sex and ethnicity, and assist in evaluating and guiding public policies

regarding the prevention and handling of this important public health problem.
C EP

Methods
AC

This is a population study that uses descriptive statistics to characterize the

demographic, social and economic profile of people suffering from TBI in Brazil. Request for

the approval of the Ethics Committee in Research for the use of the data was not necessary

because it is freely accessible and its use implies no risk to the well-being of the population

studied.
 ACCEPTED MANUSCRIPT
The information used to analyze the profile of hospital TBI in Brazil is available on

the database of the Information Technology Department of the Unified Health System

(DATASUS), available to the public online at www.datasus.com.br. The database consulted is

fed by the filling out the form called “hospital admission authorization (AIH)” by the public

and private health institutions that make up the Unified Health System (SUS). Analysis of the

PT
data available from January 2008 to December 2012 was performed. Based on the

International Disease Classification, 10th Revision (ICD-10), the terms “skull and facial bone

RI
fracture” and “head injury” were selected from a list of diagnoses. These two diagnostic

SC
terms were related to head injury among the options available for the research, and as such,

these terms were chosen based on the WHO recommendations.(22)

Using the research tool

U
available on the DATASUS website
AN
(http://www2.datasus.gov.br/DATASUS/index.php?area=0203), the diagnoses associated with
M

TBI were listed with the number of admissions, total cost of the admissions, average cost per

admission, number of deaths and the case-fatality rate, i.e. the proportion of deaths among
D

hospitalized patients. These data were further discriminated by sex, age group, ethnicity, year
TE

and region of occurrence. Other indices, such as the incidence (number of new cases in the

population per year), specific mortality (number of deaths in the population per year) and
EP

incidence in the population adjusted by age, were derived from the information of the total
C

resident population and age distribution of the population for each year. For research on the
AC

general incidence of admissions due to TBI since 1984, a database using ICD-9 as the

reference for the classification of diseases was analyzed, which was available for research

from 1984 to 1997. To find cases related to TBI in this last instance, terms available only for

this period were used: “Fracture of Skull and Face”, “Concussion”, “Intracranial injury,

except that associated with fracture of skull and concussion” and “Injury of nerves and optic

pathways and other cranial nerves ”.

 ACCEPTED MANUSCRIPT

Results

Analysis of the available DATASUS data from 2008 to 2012 showed that there were

around 125,500 hospital admissions a year due to TBI in Brazil. The incidence was 65.7

PT
admissions per 100,000 inhabitants per year, considering a population of 191 million

inhabitants, calculated based on a DATASUS estimate for the Brazilian population from 2008

RI
to 2012 (Table 1).

Only data relative to in-hospital mortality due to TBI are available on the DATASUS

SC
database, not pre-admission mortality. On average, 9,715 deaths occurred annually in patients

U
admitted due to TBI, with a hospital mortality of around 5.1/100,000/year, and a case-fatality
AN
rate of 7.7%. The average annual all-in costs of hospital expenses for patients suffering from

TBI was around R$156,300,000 (US$ 70,960,000) with an average cost per admission of
M

R$1,235.00 (US$568). The average hospitalization time was 5.5 days.

In absolute numbers, the age group 20-29 was the most frequently admitted to
D

hospital because of TBI and presented the largest number of hospital deaths. However, when
TE

relating these as percentages of the resident population per age group, the population over 80
EP

had the highest admission rate due to TBI, with around 138/100,000/year, followed by the

age group 70-79 (92.5/100,000/year), with 20-29 following in third (83/100,000/year) (Table
C

2).
AC

Men were hospitalized almost 3.5 times more frequently for TBI than women, such

that the incidence of TBI in the male population was 102/100,000/year. Male sex was also

associated with higher hospital expenses and longer hospitalization time, in addition to

greater in-hospital case-fatality.

 ACCEPTED MANUSCRIPT
Table 2 also presents case distribution by race and ethnicity of patients admitted

suffering from TBI. Unfortunately, a large number of cases exist for which this data was

absent, which hampers comparisons of the admission data for race/ethnicity.

Figure 1 shows the estimated incidence of cases of TBI since 1984, showing

reasonable oscillation from 1984 to 1998, with a peak of almost 100 cases per 100,000

PT
inhabitants in 1991, and stability from 1999 to 2012.

RI
Discussion

SC
The primary objective of this study was to provide nationwide estimates of the

U
epidemiological, social and economic profile of people suffering from TBI in Brazil. The data
AN
were analyzed based on DATASUS, a database maintained by the Ministry of Health, for

which the input is provided by filling out the hospitalization form in the Brazilian public
M

health system (SUS). It is estimated that around 80% of the Brazilian population depends

exclusively on the SUS for access to health services.(19) In cases of potentially grave injury
D

treated by public mobile emergency services, the tendency is to transport the patient initially
TE

to a public hospital and, thereafter to transfer them to the institution covered by personal
EP

health insurance plans or other private health institution, assuming the patient has health

insurance or chooses to pay for their admission. Thus, although there is no reliable
C

information on the participation of the public system in the treatment of TBI patients, we
AC

estimate that, in these cases, coverage by the public health system would amount to at least

80%.

In our study, we estimate that around 125,500 admissions of patients with TBI occur

per year in Brazil, corresponding to an incidence of 65.5/100,000 inhabitants/year, 9,700

hospital deaths due to TBI a year, with a hospital mortality coefficient of around 5/100,000

inhabitants a year. It was not possible to access the pre-hospital mortality of TBI cases.
 ACCEPTED MANUSCRIPT
Fernandes and Silva(17) also studied the epidemiology of TBI in Brazil, in the 14 to 69 age

range from 2000 to 2007, based on data provided by DATASUS, having verified an average

of 62,800 admissions a year, with an incidence of around 37/100,000 inhabitants in the age

range studied. Table 3 shows the main epidemiological characteristics reported in some

studies from different countries. When compared with the results of international studies, the

PT
incidence of admissions due to TBI in Brazil appears to be underestimated. It is reasonable to

RI
infer that in a less developed country, with low observance and enforcement of traffic

regulations, high indices of violence and inadequate infrastructure, a higher incidence of TBI

SC
would exist in relation to developed countries like the USA and European countries; however,

this is not what the estimates of the majority of Brazilian studies show. The difference in the

U
incidences determined can likely be explained by social/economic factors, methodological
AN
differences, different policies regarding hospitalization in each institutional region and
M

inadequate filling out of the AIH form or even its absence.

It is important to emphasize that that the inclusion of emergency department visits,

D

hospital admissions, and pre-hospital deaths in a study is recommended to better evaluate the
TE

epidemiological profile of TBI. In 2009, the CDC estimated the occurrence of around 2

million cases of emergency department visits in the USA (686/100,000)(4). The DATASUS
EP

database does not take into account emergency department visits, only admissions, and does
C

not include possible victims of TBI that die before arriving at hospital. Moreover, the
AC

majority of international studies on the epidemiology of TBI do not clearly discriminate

between cases of emergency department visits, admissions and pre-hospital deaths, and this

seems to be a contributing factor to the wide variation in the incidence and mortality due to

TBI reported in different studies.

The research tool of the DATASUS website does not supply information that allows

us to classify the causes of admissions with TBI in Brazil.

 ACCEPTED MANUSCRIPT
It is possible to examine deaths through external causes on the DATASUS database,

but the causes do not include the event that led to death (car accident, fall, physical

aggression, etc), and it was not possible for authors to discriminate cases where TBI was

present. However in the work of Fernandes and Silva(17), who also analyzed TBI nationwide

using the data provided by DATASUS, it was possible to classify the causes of admissions

PT
due to TBI. According to the authors, the main causes were falls (35%), followed by traffic

RI
accidents (31%) and assaults (8%). This work also highlights the increase in admissions with

TBI caused by motorcycle accidents, from 2,749 in 2001 to 7,574 in 2007, an increase of

SC
175%. Regarding other Brazilian estimates, a study in the city of Florianópolis, SC, involving

176 patients with TBI, traffic accidents were shown to be the main cause of admission

U
(41.5%), followed by falls (30.4%).(9) Another study involving 555 patients in the city of
AN
Salvador, BA, showed that traffic accidents were the main cause of TBI (41%), followed by
M

violence (25%) and falls (24%).(11) The CDC estimated that falls are the main cause both of

attendance by the emergency services and of hospital admissions, but traffic accidents are the
D

main cause of death due to TBI in the USA.(4)

TE

Very few epidemiological studies in Brazil address the severity of TBI and in the

majority, only hospitalized patients are registered. Based on the Glasgow Coma Scale (CGS),
EP

Melo et al. verified a distribution of 38.4% mild (GCS 13-15), 23.4% moderate (GCS 9-12),
C

and 38.2% severe (GCS 3-8) TBI in 380 patients admitted to a trauma reference center
AC

hospital in Salvador city.(12) In 101 patients admitted to a trauma reference center in the city

of Petrolina, PE, the reported severity distribution was 53.47% mild, 25.73% moderate and

20.8% severe TBI.(14) Masini studied a sample of 100 patients admitted to the emergency

department of the Hospital de Base de Brasília in 1991, and verified a distribution of 76%

mild (GCS 15), 12% moderate (GCS 9-14), and 12% severe TBI (GCS 3-8).(8)
 ACCEPTED MANUSCRIPT
Classification of TBI severity based on GCS alone may have its limitations, and

additional criteria can be used to enhance the correlation between classification and

prognosis. It is suggested that the duration of loss of consciousness, altered consciousness

and post-traumatic amnesia, as well as the Abbreviated Injury Severity Scale be used along

with GCS to evaluate injury severity.(3) However, severity stratification of TBI is a

PT
challenging task in larger population studies. To characterize the impact of TBI on the

RI
population more accurately it is necessary to know the general mortality rate (hospital and

pre-hospital), which is not available for consultation in the database analyzed. It is possible,

SC
however, to roughly estimate the general mortality rate due to TBI in Brazil in accordance

with the general mortality rate through external causes, extrapolating from American

U
literature that around 30% of the deaths due to external causes presented a diagnosis of
AN
TBI.(18) From 2008 to 2011 (the period available for consultation of mortality due to external
M

causes), there were on average 140,900 deaths per year due to external causes. An estimate of

around 42,280 deaths (including pre-hospital deaths) due to TBI can be made for this period,
D

with a mortality rate of 22.1/100,000/year.

TE

The high incidence of TBI verified in patients over 70 years of age, and particularly in

people over 80, should be highlighted. Although the age group most frequently affected in
EP

this study was that of 20-29, when taking the populations of the age groups into account, the
C

greatest incidence was determined for the elderly population. There is also evidence of a high
AC

morbimortality related to this injury in the elderly, with increasing case-fatality rates from the

age group 50-60 onward (13.5%), culminating in a case-fatality rate of 20% in the over 80s

age group. This finding is corroborated by other studies, which reinforces the vulnerability of

the geriatric age bracket to TBI(18,21) and can be explained by the increase in life expectancy

associated with an increase in the mobility of the elderly. It can also signify a lack of family

support, infrastructure and adequate assistance policies for these patients, who become
 ACCEPTED MANUSCRIPT
dependent for a large part of their daily activities. Falls are the main cause of TBI in the

elderly (51%), followed by motor vehicle accidents.(21) In 2003 in the USA, the total cost of

admissions of elderly patients with TBI was estimated at more than US$2.2 billion, and if the

population of the elderly in the USA doubles from the present 35 million to 70 million, as

foreseen for 2040, the costs of this injury in the elderly will be alarming.

PT
In terms of the social/economic cost, TBI also stands out as an important public health

RI
problem. Our study estimates the annual direct costs of admissions for TBI in Brazil at

around R$156,300,000 (US$ 70,960,000), and an average cost per admission of R$1,250

SC
(US$568). Studies conducted in the USA show that when direct and indirect medical costs,

loss of workforce and the negative impact on the quality of life are factored in, the

U
expenditure is of the order of US$60.4 to US$221 billion.(16,18) Another study based on the
AN
Thomason Reuters MarketScan Research Database estimated the median cost of an
M

admission for TBI in the USA at around US$10,700, with variation in the 20% to 80%

percentiles from US$5,560 to US$24,600,(6) a much higher value than that estimated for the
D

admission of a patient for TBI in Brazil. This important discrepancy can be explained by the
TE

financing of health through a public system in Brazil, which implies lower admission costs, in

contrast to the predominance of a system of private health funding in the USA, and the
EP

employment of advanced material and human resources technology, making admission costs
C

much more expensive.

AC

The importance of TBI as a main cause of morbimortality in an eminently young

population should be emphasized. It is the most common cause of disability in the young

population, this type of injury frequently results in lasting motor, cognitive, behavioral and

emotional changes that interfere in several aspects of the daily routine for the rest of the

individual’s life.(16) It is estimated that in the USA, 3.2 to 5 million people live with

limitations related to TBI.(1,4) Data in the literature on the prevalence of TBI in Brazil were
 ACCEPTED MANUSCRIPT
not found. In an attempt to roughly estimate the prevalence of individuals with limitations

related to TBI in Brazil, we used the epidemiological concept of prevalence, in which the

prevalence of a disease in the population is equal to its incidence multiplied by the time of

duration. Based on the data contained in the American literature, we calculated that the

average duration of a limitation related to TBI is 17 years, assuming an incidence of 300,000

PT
hospital admissions per year in 2009 and a previously estimated prevalence of around 5

RI
million individuals.(1) Thus, we estimate that there are around 2,130,000 people living with

limitations related to TBI in Brazil. The indirect costs of TBI, i.e. the treatment and

SC
rehabilitation of these individuals, are not included in this estimate.

Epidemiological vigilance in relation to the neurotrauma that is practiced in Brazil

U
and in the majority of countries is incomplete or inexistent. Although the effort of the Public
AN
Health System in Brazil to develop and maintain a wide-ranging database should be
M

recognized, special attention is required in the case of neurotrauma because of the

considerable impact on the quality of life of the individuals and the costs to the government.
D

The research mechanism available on the DATASUS website does not enable adequate
TE

estimation of this important problem in public health. It is not possible to estimate the general

incidence of TBI because there is no requirement to register cases dealt with by the
EP

emergency sector that do not involve hospital admission, pre-hospital deaths or TBI cases
C

that were dealt with in the outpatient sector or in doctors’ surgeries. Neither was it possible to
AC

classify the causes of TBI in Brazil, an essential element in the development of primary

prevention policies. It is still not possible to reliably estimate the total costs of treatment,

rehabilitation and care of the patients suffering from brain injury.

Conclusion
 ACCEPTED MANUSCRIPT
TBI should be recognized as an important public health problem in Brazil, responsible

for 125,500 admissions a year and 9,700 hospital deaths. It is not yet possible to properly

estimate pre-hospital mortality. The presence of long-lasting disability caused by TBI should

be better evaluated in studies in Brazil and emerging countries, because it generates

considerable onus for society.

PT
Besides the young adult age group (20-29 years old), the geriatric age group is

RI
especially vulnerable to TBI.

The implementation of a system of effective epidemiological vigilance is urgent to

SC
more fully understand the impact of TBI on the Brazilian population and worldwide, and to

assist in the implementation of primary prevention strategies and the reduction of the

morbimortality and costs of TBI.

U
AN
M

Acknowledgments

In memory of Professor Atos Alves de Souza, who founded and guided the practice of
D

neurosurgery in the Santa Casa de Belo Horizonte Hospital.

TE
EP

Conflict of Interest Statement

C

The authors declare that the article content was composed in the absence of any
AC

commercial or financial relationships that could be construed as a potential conflict of

interest.
 ACCEPTED MANUSCRIPT
References

1. Agrawal A, Galwankar S, Kapil V, Coronado V, Basavaraju SV, McGuire LC, Joshi R,

Quazi SZ, Dwivedi S. Epidemiology and clinical characteristics of traumatic brain

injuries in a rural setting in Maharashtra, India. 2007-2009. Int J Crit Illn Inj Sci

PT
2:167-171, 2012

2. Bener A, Omar AO, Ahmad AE, Al-Mulla FH, Abdul Rahman YS. The pattern of

RI
traumatic brain injuries: a country undergoing rapid development. Brain Inj 24:74-

80, 2010

SC
3. Brasure M, Lamberty GJ, Sayer NA, Nelson NW, MacDonald R, Ouellette J, Tacklind

U
J, Grove M, Rutks IR, Butler ME, Kane RL, Wilt TJ. Multidisciplinary Postacute
AN
Rehabilitation for Moderate to Severe Traumatic Brain Injury in Adults, in. Rockville:

Agency for Healthcare Research and Quality, 2012

M

4. Coronado VG, McGuire LC, Sarmiento K, Bell J, Lionbarger MR, Jones CD, Geller

AI, Khoury N, Xu L. Trends in Traumatic Brain Injury in the U.S. and the public
D

health response: 1995-2009. J Safety Res 43:299-307, 2012

TE

5. Feigin VL, Theadom A, Barker-Collo S, Starkey NJ, McPherson K, Kahan M, Dowell

EP

A, Brown P, Parag V, Kydd R, Jones K, Jones A, Ameratunga S, Group BS. Incidence

of traumatic brain injury in New Zealand: a population-based study. Lancet Neurol

C

12:53-64, 2013
AC

6. Hu J, Ugiliweneza B, Meyer K, Lad SP, Boakye M. Trend and geographic analysis for

traumatic brain injury mortality and cost based on MarketScan database. J

Neurotrauma 30:1755-1761, 2013

7. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of

traumatic brain injury: a brief overview. J Head Trauma Rehabil 21:375-378, 2006
 ACCEPTED MANUSCRIPT
8. Masini M. Epidemiological profile of craniocerebral injuries in the Federal District in

1991. J bras neurocir 5:61-68, 1991

9. Martins ET, Linhares MN, Sousa DS, Schroeder HK, Meinerz J, Rigo LA, Bertotti

MM, Gullo J, Hohl A, Dal-Pizzol F, Walz R. Mortality in severe traumatic brain

injury: a multivariated analysis of 748 Brazilian patients from Florianopolis City. J

PT
Trauma 67:85-90, 2009

RI
10. Maset A, Andrade A, Martucci SC, Frederico LM. Epidemiologic features of head

injury in Brazil. Arq bras neurocir 12:293-302, 1993

SC
11. Melo JRT, Silva RA, Moreira-Junior ED. Characteristics of Patients with Traumatic

Brain Injury in the City of Salvador, Bahia, Brazil. Arq Neuropsiquiatr 62:711-715,

2004
U
AN
12. Melo JRT, Oliveira-Filho J, Silva RA, Moreira-Junior ED. Prognostic factors about
M

mortality and lethality in head injury. Arq Neuropsiquiatr 63:1054-1057, 2005

13. Menon DK, Schwab K, Wright DW, Maas AI; Demographics and Clinical Assessment
D

Working Group of the International and Interagency Initiative toward Common Data
TE

Elements for Research on Traumatic Brain Injury and Psychological Health. Position

statement: definition of traumatic brain injury. Arch Phys Med Rehabil 91:1637-
EP

1640, 2010
C

14. Moura JC, Rangel BLR, Creoncio SCE, Pernambuco JRB. Clinical and
AC

epidemiological profile of head trauma at the Urgências e Traumas Hospital in the

city of Petrolina, Pernambuco. Arq Bras Neurocir 30:99-104, 2011

15. Nell V, Brown DS. Epidemiology of traumatic brain injury in Johannesburg--II.

Morbidity, mortality and etiology. Soc Sci Med 33:289-296, 1991

16. Ponsford JL, Spitz G, Cromarty F, Gifford D, Attwood D. Costs of care after

traumatic brain injury. J Neurotrauma 30:1498-1505, 2013

 ACCEPTED MANUSCRIPT
17. Fernandes RNR, Silva M. Epidemiology of Traumatic Brain Injury in Brazil. Arq

Bras Neurocir 32:136-142, 2013

18. Roozenbeek B, Maas AI, Menon DK. Changing patterns in the epidemiology of

traumatic brain injury. Nat Rev Neurol 9:231-236, 2013

19. Ministry of Health of Brazil. Executive Secretariat. More Health : a right for everyone

PT
: 2008 – 2011, in Editora do Ministério da Saúde (ed). Brasília, 2008

RI
20. Tagliaferri F, Compagnone C, Korsic M, Servadei F, Kraus J. A systematic review of

brain injury epidemiology in Europe. Acta Neurochir (Wien) 148:255-268;

SC
discussion 268, 2006

21. Thompson HJ, McCormick WC, Kagan SH. Traumatic brain injury in older adults:

U
epidemiology, outcomes, and future implications. J Am Geriatr Soc 54:1590-1595,
AN
2006
M

22. Thruman DJ, Holder Y. Uniform Standards for Neurotrauma Surveillance, in

Thruman DJ, Kraus JF, and Romer CJ. (ed): Standards for Surveillance of
D

Neurotrauma. Switzerland: World Healh Organization, 1995, pp 9-39

TE

23. Wright RM, Post A, Hoshizaki B, Ramesh KT. A multiscale computational approach

to estimating axonal damage under inertial loading of the head. J Neurotrauma

EP

30:102-118, 2013
C
AC
 ACCEPTED MANUSCRIPT

Table 1. Epidemiological aspects of traumatic brain injury in Brazil, 2008-2012.

2008 2009 2010 2011 2012 Mean

PT
Hospital Admissions 106,695 125,476 131,326 133,085 131,175 125,551

RI
Incidence (/100,000 pop.) 57.2 66.5 68.8 68.9 67.1 65.7

SC
Total Cost (US$) 47,988,979.09 66,660,779.41 74,890,240.91 81,114,896.09 84,709,145.91 71,072,808.18

Mean Cost per Admission

U
(US$) 449.55 531.26 570.26 609.50 645.31 561.36

AN
Mean Length of Hospital
Stay (days) 5.4 5.3 5.5 5.6 5.8 5.5

M
Hospital Deaths 8,378 9,790 10,593 10,051 9,761 9,715

D
In-Hospital Mortality

TE
(/100,000 pop.) 4.5 5.2 5.5 5.2 5.0 5.1
EP
In-Hospital Lethality (%) 7.85 7.8 8.07 7.55 7.44 7.7

Pop.= population
C
AC
 ACCEPTED MANUSCRIPT

Table 2. Epidemiological aspects of traumatic brain injury in Brazil, according to age group, sex and ethnicity.
Mean Length of
Hospital Incidence Mean Cost per Hospital Stay In-Hospital
Admissions (/100.000 pop) Total cost (US$) Admission (US$) (days) Deaths Case-Fatality (%)

PT
Age Group (years)
<1 2,166 74.5 670,500.49 309.56 3.6 54.6 2.5
1-4 5,550.8 47.1 1,542,784.98 277.94 3.1 80.8 1.5

RI
5-9 5,861.6 37.0 1,856,795.98 316.77 3.4 100.4 1.7
10 - 14 5,601.4 33.0 2,253,136.52 402.25 4.1 158.8 2.8

SC
15 - 19 11,355.2 66.7 6,380,591.27 561.91 5.2 668.0 5.9
20 - 29 28,905.4 83.0 16,740,704.24 579.15 5.4 1,770.4 6.1
30 - 39 21,009.2 71.3 12,141,128.71 577.91 5.6 1.452 6.9

U
40 - 49 16,598.8 67.3 10,245,885.09 617.27 6.2 1,407.8 8.5
50 - 59 11,404.2 63.2 7,437,097.86 652.14 6.7 1,250.6 11.0

AN
60 - 69 7,439.4 67.4 5,123,286.14 688.64 6.8 1,010.2 13.6
70 - 79 5,717.6 92.5 4,040,683.41 706.68 6.8 937.2 16.4

M
> 80 3,960.6 138.2 2,640,213.59 666.59 6.6 823.8 20.8
Sex
Male 97,552.0 102.1 58,170,260.00 596.36 5.8 7,897.0 8.1

D
Female 28,017.0 29.3 12,902,548.18 460.45 4.7 1,817.0 6.5
Ethnicity

TE
White 3,610,932 24,751,340.45 614.55 5.0 2,973.0 7.38
Black 345,778.8 2,386,002.73 637.73 6.0 302.0 8.1
EP
Mixed Race 3,142,782.4 15,311,784.09 589.09 5.6 1,765.0 6.8
Indigenous 31,984.0 83,390.00 429.82 4.7 9.4 4.8
Yellow 50,030.0 422,629.09 469.77 4.8 60.0 6.71
C

No Information 3,937,101.8 28,117,661.82 516.14 6.0 4,605 8.45

AC
 ACCEPTED MANUSCRIPT

Table 3. Comparison of the present study with other national and international studies.
Mortality In-Hospital
ED visits + ED Visits HA (general) Mortality Moderate/
5 5 5 5 5
Author Period Location HA (/10 ) (/10 ) (/10 ) (/10 ) (/10 ) Mild TBI Severe TBI MLoS (days) Inclusion criteria
Almeida et
al. (present 2008-

PT
study) 2012 Brazil - - 65.7 - 5.1 - - 5.5
Hospital Information System
Fernandes & 2000- <4 days in 49% database from Health Ministry.

RI
Silva (9) 2007 Brazil - - 37 - 4.3 - - of cases Population age 14-69 years old.

SC
Johannesburg, Trauma referral center
Nell et al. (6) 1991 South Africa - - 316 80 - 90% 10% admissions. Death certificates.

U
Trauma referral center ED visits

AN
Masini et al. and HA, in its catchment area.
(8) 1991 Brasília, Brazil 341 - - 50 - - - - Death certificates.
Trauma referral center ED visits

M
and HA, in its catchment area.
Bener et al. 2003- Study did not include TBI data
(5) 2007 Qatar 44* - - - 70% 30% - due to fighting or assault.

D
TE
Hamilton and Retrospective and prospective
Feigin et al. 2010- Waikato District, broad surveillance in health care
(21) 2011 New Zealand 790 - - - - 95% 5% - facilities. Death certificates.
EP
Trauma referral center overnight
Maset et al. 1986- São Jose do Rio hospital stay and catchment area.
C

(7) 1987 Preto, Brazil 456† - - 38 - - - 4.65 Death certificates.

Data from CDC surveillance
AC

system. Includes nationwide

Coronado et emergency department visits,
al. (3) 2009 USA 781.7 686 95.7 16.6 - - - - hospital admissions and deaths.
5
ED= Emergency Department; HA= Hospital Admissions; MLoS= Mean Length of Hospital Stay; (10 )= rates per 100,000 population; TBI= traumatic brain injury.
*Data due to fighting or assault not included in the study.
†Patients discharged less than 24h after arrival excluded.
MLoS, Median length of hospital stay.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
Abbreviations list:

TBI: traumatic brain injury

DATASUS: Information Technology Department of the Centralized Health System
CDC: Centers for Disease Control and Prevention
WHO: World Health Organization

PT
USA: United States of America
AIH: hospital admission authorization

RI
SUS: Centralized Health System (as in Portuguese: Sistema Único de Saúde)

SC
ICD-10: International Disease Classification 10th Revision

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Conflict of Interest:

All authors declare no conflict of interest.

Funding and Ethics Committee approval:

No funding or ethics committee approval was necessary in this work.

PT
Authors:

RI
Carlos Eduardo Romeu de Almeida

Jose Lopes de Sousa Filho

SC
Jules Carlos Dourado

Pollyana Anício Magalhães Gontijo

Marcos Antônio Delaretti

U
AN
Bruno Silva Costa
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Traumatic Brain Injury Epidemiology in Brazil

Highlights:

• Population over 80 showed the highest TBI admission rate, with around

PT
138/100,000/year.

RI
• Incidence of hospital TBI admissions to the Brazilian NHS is

SC
65.7/100,000/year.

• There were around 125,500 hospital admissions a year due to TBI in

Brazil.
U
AN
• Hospital mortality was 5.1/100,000/year, with a case-fatality rate of 7.7%.
M

• The average annual costs for patients treated for TBI were around US$

70,960,000.
D
TE
C EP
AC