Injury, Int. J. Care Injured 41 (2010) 913–917

The changing face of mild head injury: Temporal trends and patterns inadolescents and adults from 1997 to 2008

Andrea Fabbri a,*, Franco Servadei b, Giulio Marchesini c, Antonella Negro d, Alberto Vandelli a

a Dipartimento dell’Emergenza, Presidio Ospedaliero Morgagni-Pierantoni, Azienda Unita Sanitaria, Locale di Forlı, Forli, Italyb Unita Operativa di Neurochirurgia, Azienda Ospedaliero-Universitaria di Parma, Parma, Italyc Alma Mater Studiorum, Universita di Bologna, Bologna, Italyd Agenzia Sanitaria e Sociale Regionale, Regione Emilia Romagna, Bologna, Italy

A R T I C L E I N F O

Article history:

Accepted 1 March 2010

Keywords:

Epidemiology

Mild head injury

Outcome

Temporal trends

A B S T R A C T

Objective: To explore the temporal trend of incidence, causes of injury and main characteristics of

adolescent and adult subjects with mild head injury (MHI).

Design: This study had a retrospective design.

Setting: The study was conducted in a longitudinal database of an Italian Emergency Department (ED).

Participants: The study comprised 19 124 consecutive subjects who visited and were managed within

24 h from the event, according to a predefined protocol for MHI from 1997 to 2008.

Main outcome measures: Incidence, demography, cause of injury and characteristics of any post-

traumatic intracranial lesion within 7 days from MHI.

Results: The number of subjects with MHI decreased from 2019 per year (1997–1999) to 1232 per year

(2006–2008; P for linear trend<0.001), without differences in the total number of subjects visited in the

ED. The decrease was observed in all age-decades, in particular, in subjects in the age ranges of 20–29 and

30–39 years. Over time, the age of subjects with MHI lost a bimodal distribution, and the mean age

increased from 43 (25–69) years (median (interquartile range)) in 1997–1999 to 56 (33–78) years in

2006–2008 (P < 0.001). The prevalence of falls increased from 36.5% to 55.0%, whereas crashes fell from

53.2% to 31.9%. The incidence of subdural haematoma (SDH) and epidural haematoma (EDH) did not

change over time, whereas traumatic subarachnoid haemorrhage (t-SAH) and intra-cerebral

haematoma/brain contusion (ICH) increased (from 0.7% to 1.9% and from 2.5% to 3.2%; P for trend:

<0.001 for both.

Conclusions: The incidence and the clinical characteristics of MHI subjects are rapidly changing in our

setting. These data need to be considered in defining the effectiveness of preventive measures and

deciding resource allocation.

� 2010 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Injury

journa l homepage: www.e lsevier .com/ locate / in jury

Background

Mild traumatic brain injury (MHI) constitutes a major healthand socioeconomic problem in Western countries, accounting forhigh mortality and major disability among young individuals.16

A 1974–1975 survey showed that the highest incidence of MHIoccurred in males between 15 and 24 years and in men and womenolder than 65 years.9 This figures were confirmed in large Europeancountries between 1990 and 20009,17,21,23 as in a more recent USstudy.6 Falls and motor vehicle crashes were initially indicated as

* Corresponding author at: Dipartimento dell’Emergenza, Presidio Ospedaliero

Morgagni-Pierantoni, Azienda USL Forlı, Via Forlanini 34, I, 47100 Forli, Italy.

Tel.: +39 0543 735800; fax: +39 0543 735163.

E-mail address: dr.andrea.fabbri@gmail.com (A. Fabbri).

0020–1383/$ – see front matter � 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.injury.2010.03.002

the leading causes of injury in the two different subsets ofpopulation,8,18 with falls accounting for most head injury-relatedhospitalisations in the elderly10 and crashes responsible for MHI inmale adolescents.14

In the last few years, the incidence of MHI in developedcountries has dropped.27 This trend was suggested to derive fromchanges in the characteristics of trauma population, which hasprogressively shifted from younger towards older age groups.5

Importance

An appropriate targeting of preventive measures aimed atimproving outcome requires a detailed understanding of inci-dence, causes of injury and treatment approaches. In particular, adetailed assessment of the MHI profile is pivotal to further exploitpreventive measures in relation to a changing epidemiology.

A. Fabbri et al. / Injury, Int. J. Care Injured 41 (2010) 913–917914

Goal of study

Since 1997, a complete database of adolescent (age: �10 years)and adult subjects with head injury has been set up in our hospital.The present study aimed to analyse the main characteristics andthe temporal trends of subjects attending the Emergency Depart-ment (ED) of a General Hospital for MHI from 1997 to 2008.

Materials and methods

Selection of participants and study design

All subjects aged�10 years attending the ED of the Local HealthDistrict of Forlı were enrolled in a prospective registry for outcomepurposes. The register was set up according to the recommenda-tions of the Neurotraumatology Committee of the World Federa-tion of Neurosurgical Societies.20

Setting

The Head Injury Registry consists of over 20 000 consecutivemedical records of all subjects attending the ED of Forlı Hospitalwithin 24 h from the event (from January 1997 to December 2008).MHI was defined as an acute injury of the head, other than anysuperficial injury to the face, Glasgow Coma Scale (GCS) 14 or 15 inadolescent and/or adult inhabitants aged �10 years.

The ED of Forlı is the only referral center where emergencies aretreated and is part of a general hospital with 450 acute beds,providing general clinical services (except cardiac surgery andneurosurgery). The ED treats nearly 50 000 children and adults peryear, from a population of 171 000 inhabitants in an area of1380 km2, 58% of the population living in an urban environment.The hospital admits >20 000 subjects, with 900 admissions to theintensive care unit. The hospital database is directly connectedwith the General Registry Office of the District.

Excluded from the present analysis were 3366 subjects (17.6%)because of (a) unclear history of injury as the primary event (1388cases; 0.7%); (b) unstable vital signs (580 cases; 0.3%); (c) GCS< 14(382 cases; 0.2%); (d) penetrating injuries at presentation (42 cases;0.1%); (e) discharge against medical advice (573 cases; 0.3%) and (f)duplicate records of subjects re-attending for post-concussivesyndrome within 7 days of admission for MHI (402 cases; 2.0%).

In addition, a systematic analysis of the outcome of all subjectswith documented intracranial lesions (index cases) was carried outby checking the medical databases of our local health district.Following this, a member of the ED staff contacted all survivingindex cases by means of a structured telephonic interview, toevaluate the GCS at the 6-month follow-up.

Treatment protocol

Subjects were treated according to a predefined diagnostic andobservational protocol as previously published.4 Computedtomography (CT) was not mandatory at presentation, but allsubjects re-attending the ED for trauma-related symptoms werestudied with CT scans.

CT scans showing intracranial lesions were urgently reviewedby a neurosurgeon of the nearest Neurosurgical Unit (Cesena,17 km away), consulted via a teleradiology system. The protocolwas carried out according to the Helsinki Declaration and approvedby a Committee of the Local Health District of Forlı.

Data collection and processing

In the present analysis, the following variables were consid-ered:

1. Demography and mechanism of injury2. Clinical characteristics and disposition3. Ordering rate of head CT scans4. Type of intracranial lesion5. Need for neurosurgical evacuation for intracranial lesion6. Unfavourable outcome at the 6-month follow-up

Variable definition and outcome measures

The main outcome measure was a diagnosis of any post-traumatic lesion at CT within 7 days from trauma. Lesions included(a) intra-cerebral haematoma/brain contusion(s) (ICH); (b) trau-matic subarachnoid haemorrhage (t-SAH); (c) subdural haema-toma (SDH) and (d) epidural haematoma (EDH).22

In subjects revisiting for complaints, a time limit of 7 days wasset to consider any lesion as directly related to the initial headinjury. Similarly, haematoma evacuation and/or skull fractureelevation were only considered within 7 days after diagnosis.

The 6-month outcome was considered unfavourable if thepatient died of head injury-related causes, remained in avegetative state or had a permanent severe disability, measuredby the Glasgow Outcome Scale (GOS).25 GOS was rated during thestructured telephone call up to June 2009.

Analyses

Mean value, standard deviation (SD) and frequencies were usedto describe data distribution. To improve clarity of analyses andpresentation, all data collected during 12 years were grouped infour 3-year periods (1997–1999; 2000–2002; 2003–2005 and2006–2008). We tested the temporal trends and the associations ofvariables with the outcome measures using the Cochran–ArmitageTrend Test. The SPSS software, version 17.0 (SPSS Inc., Chicago, IL,USA) and the Statistical Analysis Software (SAS) software, version9.0 (SAS Institute Inc., Cary, NC, USA) were used for statisticalanalyses.

Results

Age- and sex-related trends in MHI

Between 1997 and 2008, the number of subjects attending theED of Forlı for MHI progressively decreased from 2019 per year(period 1997–1999) to 1232 per year (2006–2008), (P for trend:<0.001), without differences in the total number of subjects visitedin the ED (Table 1). Over time, subjects presenting with MHIdecreased in all decades (Fig. 1) (P for trend per decades: <0.001),particularly in the age-decades 20–29 (by 70%) and 10–19 or 30–39 years (by 50%) (Fig. 2). When compared with inhabitants in theForlı area, the ED-admission rate for MHI in the age range of 10–29decreased in the same period from 16.1 per 1000 in the year 2000to eight per 1000 in 2007; in the same period, the admission ratealso decreased in the age group of 60–70 years from seven per 1000to six per 1000 inhabitants. The more pronounced decrease inyoung people produced an increase in the median age of MHIpatients from 43 years (interquartile range (IQR): 25–69) in theperiod 1997–1999 to 56 years (33–78) in the period 2006–2008 (P

for trend: <0.001) (Table 1). The increase of age was observed inboth sexes, with a male/female ratio remaining fairly constantfrom 1997 to 2008 (�60:40 in the age range of 20–30,progressively turning into 40:60 in subjects aged �70).

Age- and sex-related trends in cause of injury

Falls and motor vehicle crashes together were the main causesof MHI (88.0% of cases), each accounting for 44.0% of cases in the

Table 1Clinical characteristics of subject visited in the Emergency Department with a diagnosis of mild head injury along the years. N. cases and N. visited in ED are reported as

number/year in each period. Age is reported as median (interquartile range). Undiagnosed cases at first evaluation are reported as number of cases (%).

Year 1997–1999 2000–2002 2003–2005 2006–2008 P value for trend

N. cases 2,019 1,759 1,385 1,232 <0.001

N. visited in ED 48,709 51,041 46,962 47,087 0.121

Males (%) 54.1 56.4 55.4 52.6 0.226

Age (years) 43 (25–69) 43 (27–71) 53 (32–77) 56 (33–78) <0.001

Crashes (%) 53.2 47.7 36.5 31.9 <0.001

Falls (%) 36.5 39.6 50.6 55.0 <0.001

Assaults (%) 2.3 3.8 3.6 3.7 <0.001

CT rate (%) 35.9 49.8 61.1 54.8 <0.001

Positive CT (%) 3.9 6.8 6.4 5.5 <0.001

Undiagnosed cases (%) 24 (0.4) 9 (0.2) 4 (0.1) 10 (0.3) 0.078

Admitted (%) 31.6 25.7 17.7 15.3 <0.001

Fig. 1. Number of cases visited in the Emergency Department for mild head injury in

four time periods, divided according to age-decades. Note the disappearance of the

bimodal distribution of age along the years.

A. Fabbri et al. / Injury, Int. J. Care Injured 41 (2010) 913–917 915

whole period. Along with changes in age-trends, a significantchange in the cause of injury occurred. Over time, the relativeproportion of falls increased from 36.5% in the period 1997–1999to 55.0% in 2006–2008, whereas crashes fell from 53.2% to 31.9%within the same periods. This was mainly related with aremarkable decrease of crash as cause of MHI, particularly in

Fig. 2. Temporal trends of subjects visited in the Emergency department for mild

head injury in four time periods (1997–1999, 2000–2002, 2003–2005, 2006–2008)

in relation to age-decades.

males (Fig. 3). The total number of falls remained relatively stable,thus increasing over time from 45.3% to 65.8% of total as cause ofinjury in male subjects (P for trend: <0.001) and from 29.1% to45.3% in females (P for trend: <0.001). The number of assaults ascause of injury was low (only 3.3% of cases), but increased overtime from 2.3% to 3.7% (P for trend: <0.001).

Trends in type of intracranial lesions

The CT ordering rate increased to �50% of cases after the year2000, in relation to a modified diagnostic protocol, with consensusamong physicians of the ED, and improved CT-scan facilities(Table 1).

Over time, intracranial lesions were diagnosed in 4.8% ofsubjects presenting to the ED for MHI in the period 1997–1999,increased to 6.8% in the period 2000–2002, and declined thereafterto 5.5% in the last period. The incidence of SDH and EDH lesions didnot change over time, with the exception of increased t-SAH (from0.7% to 1.9%; P for trend:<0.001) and ICH (from 2.5% to 3.2%) (P fortrend: <0.001). An intracranial lesion remained undiagnosed afterfirst evaluation in 47 cases (0.2% of total cases) without trend overtime (P for trend: 0.078). Only 18/47 (0.1%) subjects weredischarged with undiagnosed intracranial lesions. The incidenceof these cases did not change from 1997 to 2008 (P for trend: 0.545)

The need for urgent neurosurgical intervention did not changeover time (N = 177; 0.8% of cases, P = 0.793), and an unfavourable

Fig. 3. Number of subjects visited in the Emergency department for mild head injury

in four time periods according to trends in causes of injury and sex.

A. Fabbri et al. / Injury, Int. J. Care Injured 41 (2010) 913–917916

outcome at 6 months was observed in 84 cases (0.4%) with nosignificant time trend (P = 0.122).

Discussion

This study shows that important changes in head injury trendsoccur over time across age, sex and the main clinical characteristicsof injury. A better knowledge of the changing profile of injuredsubjects may assist clinicians in planning different diagnosticalgorithms and resource allocation, and to derive data on theeffectiveness of preventive measures.

In our population, fewer and fewer subjects were visited forMHI over time, with changing clinical characteristics in thepopulation from younger males to older females and a shift fromcrash to falls as the main cause of injury. In the same period, thenumber of admitted subjects decreased, in agreement with similartrends observed in most European countries, owing to a wide-spread diffusion of CT-scan facilities.21

The most relevant observation drawn from our analysis is thelarge reduction in the total and relative number of young subjectsattending the ED for MHI. This translates into a lower total numberof MHI in our area. Early epidemiological studies reported that thehighest rates of injury occurred in male subjects, particularly in theage range between 16 and 20 years in both sexes, with a secondmode beyond the age of 65 years.17 However, a strict association ispresent between demographic data and the causes of injury, andthe clinical characteristics are different when analysed from ageographic point of view. For instance, motor vehicle-related headinjury was the leading cause of injury in the USA and otherEuropean countries, but not in Scandinavian countries.2,3,7,11–

13,24,26 Males were more likely to suffer from MHI than females atall ages.17 In our experience, the number of young people wasreduced by �60–70% along the years, without any remarkablechange in the number of the elderly people, at variance with dataobserved in studies from Finland.10,15

The increased median age of injured subjects is merely themathematical product of the reduced number of young peopleattending the ED with MHI along the years, in the absence ofsignificant changes in admission policies and transfers to Level 2structures. In turn, this could be correlated with the effectivenessof preventive measures in terms of road crashes, considering thatcrashes account for the large majority of MHI in this age and sexgroup. Time trends in the number of crashes in Italy show aconsistent decline in the last few years. The adoption of a nationallaw in making the use of approved helmets compulsory formotorcycles and motorbike riders and passengers in March 2000increased helmet use in our area from 20% to 96% further decreasedthe incidence of traumatic brain injuries.19 The contribution ofother preventive measures (e.g., car design, traffic enforcementand drink driving) is more difficult to disentangle. Of note, theprogressive ageing of the Italian population did not produce anabsolute increase in the number of MHI in the elderly,notwithstanding the high percentage of falls observed in ourpopulation.1 Prevention campaigns addressing the importance offalls in the elderly population are needed.

Limitations

Data collection and analysis were obtained from a single Level 1ED, and are thus prone to individual variation and clinicalinterpretation bias. The data were not specifically collected toanswer the questions addressed in this study, and potentialconfounding variables were not considered. Co-morbidities, poly-pharmacy and poor pre-injury functional status could beimportant to consider for outcome purposes, but are not expectedto bias admission rates. These data should be put into the context

of changes in the demographic characteristics of the general Italianpopulation, but the extent of the reduction of admission rates inyoung people and crash as cause of admission largely exceeds thereduction of adolescent and young adult population in the Forlıarea.

Conclusions

The profile of subjects with MHI no longer presents a bimodalage distribution in our setting. This is in agreement with dataobserved in developed countries, where prevention campaigns onroad crashes are producing the expected results. The burden ofinjury in younger people, particularly in males, although reduced,remains high considering late events and disabilities, and everyeffort should be made to further exploit preventive measures onroad crashes. These favourable results in young people should beaccompanied by campaigns to address the social problem of falls inthe elderly, which has not changed along the years.

Conflict of interest

All authors warrant to have no conflict of interest in connectionwith this article, they have access to all data in the study and theyheld final responsibility for the decision to submit for publication.

Contributors

A.F. conceived the study, wrote the protocol, coordinated thedata collection, interpretation of results and wrote the article. F.S.contributed to interpretation of the results and critical review ofthe article. G.M. contributed to study design, interpretation of theresults and co-wrote the article. A.N. contributed to statisticalanalyses and interpretation of the results. A.V. contributed tointerpretation of the results and critical review of the article. Allauthors approved the final version of the article.

Funding

None specified.

Acknowledgements

We are grateful to the Dr. Fernanda Tagliaferri U.O. UnitaOperativa di Neurochirurgia, Azienda Ospedaliero-Universitaria diParma, Italy for helpful support and critical review of themanuscript.

References

1. Butcher I, McHugh GS, Lu J, et al. Prognostic value of cause of injury in traumaticbrain injury: results from the IMPACT study. J Neurotrauma 2007;24:281–6.

2. Carroll LJ, Cassidy JD, Peloso PM, et al. Systematic search and review proce-dures: results of the WHO Collaborating Centre Task Force on Mild TraumaticBrain Injury. J Rehabil Med 2004;(43 suppl.):11–4.

3. Engberg A, Teasdale TW. Traumatic brain injury in children in Denmark: anational 15-year study. Eur J Epidemiol 1998;14:165–73.

4. Fabbri A, Servadei F, Marchesini G, et al. Prospective validation of a proposal fordiagnosis and management of patients attending the emergency departmentfor mild head injury. J Neurol Neurosurg Psychiatry 2004;75:410–6.

5. Giannoudis PV, Harwood PJ, Court-Brown C, Pape HC. Severe and multipletrauma in older patients: incidence and mortality. J Trauma 2009 [EpubFebruary 11, 2009, ahead of print].

6. Holmes JF, Hendey GW, Oman JA, et al. Epidemiology of blunt head injuryvictims undergoing ED cranial computed tomographic scanning. Am J EmergMed 2006;24:167–73.

7. Ingebrigtsen T, Mortensen K, Romner B. The epidemiology of hospital-referredhead injury in northern Norway. Neuroepidemiology 1998;17:139–46.

8. Jager TE, Weiss HB, Coben JH, Pepe PE. Traumatic brain injuries evaluated in U.S.emergency departments, 1992–1994. Acad Emerg Med 2000;7:134–40.

9. Jennett B, MacMillan R. Epidemiology of head injury. Br Med J (Clin Res Ed)1981;282:101–4.

A. Fabbri et al. / Injury, Int. J. Care Injured 41 (2010) 913–917 917

10. Kannus P, Palvanen M, Niemi S, et al. Increasing number and incidence of fall-induced severe head injuries in older adults: nationwide statistics in Finland in1970–1995 and prediction for the future. Am J Epidemiol 1999;149:143–50.

11. Kraus JF, Black MA, Hessol N, et al. The incidence of acute brain injury and seriousimpairment in a defined population. Am J Epidemiol 1984;119:186–201.

12. Kraus JF, Fife D, Conroy C. Incidence, severity, and outcomes of brain injuriesinvolving bicycles. Am J Public Health 1987;77:76–8.

13. Kraus JF, Nourjah P. The epidemiology of mild, uncomplicated brain injury. JTrauma 1988;28:1637–43.

14. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact oftraumatic brain injury: a brief overview. J Head Trauma Rehabil 2006;21:375–8.

15. Luukinen H, Herala M, Koski K, et al. Rapid increase of fall-related severe headinjuries with age among older people: a population-based study. J Am GeriatrSoc 1999;47:1451–2.

16. Maas AI, Stocchetti N, Bullock R. Moderate and severe traumatic brain injury inadults. Lancet Neurol 2008;7:728–41.

17. Peloso PM, von Holst H, Borg J. Mild traumatic brain injuries presenting toSwedish hospitals in 1987–2000. J Rehabil Med 2004;22–7.

18. Rutland-Brown W, Langlois JA, Thomas KE, Xi YL. Incidence of traumatic braininjury in the United States, 2003. J Head Trauma Rehabil 2006;21:544–8.

19. Servadei F, Begliomini C, Gardini E, et al. Effect of Italy’s motorcycle helmet lawon traumatic brain injuries. Inj Prev 2003;9:257–60.

20. Stein SC, Fabbri A, Servadei F, Glick HA. A critical comparison of clinical decisioninstruments for computed tomographic scanning in mild closed traumaticbrain injury in adolescents and adults. Ann Emerg Med 2009;53:180–8.

21. Steudel WI, Cortbus F, Schwerdtfeger K. Epidemiology and prevention of fatalhead injuries in Germany—trends and the impact of the reunification. ActaNeurochir (Wien) 2005;147:231–42 [discussion 42].

22. Stiell IG, Wells GA, Vandemheen K, et al. The Canadian CT Head Rule for patientswith minor head injury. Lancet 2001;357:1391–6.

23. Tagliaferri F, Compagnone C, Korsic M, et al. A systematic review of brain injuryepidemiology in Europe. Acta Neurochir (Wien) 2006;148:255–68 [discussion68].

24. Tate RL, McDonald S, Lulham JM. Incidence of hospital-treated traumatic braininjury in an Australian community. Aust N Z J Public Health 1998;22:419–23.

25. Thornhill S, Teasdale GM, Murray GD, et al. Disability in young people andadults one year after head injury: prospective cohort study. BMJ 2000;320:1631–5.

26. Tiret L, Hausherr E, Thicoipe M, et al. The epidemiology of head trauma inAquitaine (France), 1986: a community-based study of hospital admissions anddeaths. Int J Epidemiol 1990;19:133–40.

27. United Nations NU, New York and Geneva, Economic Commission for Europe,Geneva. ec.europa.eu/health-eu/doc/alcoholineu_summary_en.pdf Statistics ofRoad Traffic Accidents in Europe and North America 2007; LI.