Methodological Issues and Research Recommendations for Prognosis After Mild Traumatic Brain Injury:...

edicine and Rehabilitation
Archives of Physical M journal homepage: www.archives-pmr.org
Archives of Physical Medicine and Rehabilitation 2014;95(3 Suppl 2):S265-77

SPECIAL COMMUNICATION

Methodological Issues and Research Recommendationsfor Prognosis After Mild Traumatic Brain Injury:Results of the International Collaboration on MildTraumatic Brain Injury Prognosis

Vicki L. Kristman, PhD,a,b,c,d Jorgen Borg, MD, PhD,e Alison K. Godbolt, MBChB, MD,e

L. Rachid Salmi, MD, PhD,f,g,h Carol Cancelliere, DC, MPH,i,j Linda J. Carroll, PhD,k

Lena W. Holm, DrMedSc,l Catharina Nygren-de Boussard, MD, PhD,e Jan Hartvigsen, PhD,m

Uko Abara, BHSc,a James Donovan, DC,i J. David Cassidy, PhD, DrMedScd,i,j,m

From the aDepartment of Health Sciences, Lakehead University, and bDivision of Human Sciences, Northern Ontario School of Medicine, LakeheadUniversity, Thunder Bay, Ontario, Canada; cInstitute for Work and Health, Toronto, Ontario, Canada; dDivision of Epidemiology, Dalla Lana School ofPublic Health, University of Toronto, Toronto, Ontario, Canada; eDepartment of Clinical Sciences, Rehabilitation Medicine, Karolinska Institutet,Danderyd University Hospital, Stockholm, Sweden; fUniv. Bordeaux, ISPED, Centre INSERM U897-Epidemiologie-Biostatistique, F-33000 Bordeaux,France; gINSERM, ISPED, Centre INSERM U897-Epidemiologie-Biostatistique, F-33000 Bordeaux, France; hCHU de Bordeaux, Pole de sante publique,Service d’information medicale, F-33000 Bordeaux, France; iDivision of Health Care and Outcomes Research, Toronto Western Research Institute,University Health Network, University of Toronto, Toronto, Ontario, Canada; jInstitute of Health Policy, Management and Evaluation, University ofToronto, Toronto, Ontario, Canada; kSchool of Public Health and Alberta Centre for Injury Control and Research, University of Alberta, Edmonton,Alberta, Canada; lDivision of Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden; and mInstitute ofSports Science and Clinical Biomechanics, Faculty of Health, University of Southern Denmark, Odense, Denmark.

Abstract

The International Collaboration on Mild Traumatic Brain Injury (MTBI) Prognosis performed a comprehensive search and critical review of the

literature from 2001 to 2012 to update the 2002 best-evidence synthesis conducted by the World Health Organization Collaborating Centre for

Neurotrauma, Prevention, Management and Rehabilitation Task Force on the prognosis of MTBI. Of 299 relevant studies, 101 were accepted as

scientifically admissible. The methodological quality of the research literature on MTBI prognosis has not improved since the 2002 Task Force

report. There are still many methodological concerns and knowledge gaps in the literature. Here we report and make recommendations on how to

avoid methodological flaws found in prognostic studies of MTBI. Additionally, we discuss issues of MTBI definition and identify topic areas in

need of further research to advance the understanding of prognosis after MTBI. Priority research areas include but are not limited to the use of

confirmatory designs, studies of measurement validity, focus on the elderly, attention to litigation/compensation issues, the development of

validated clinical prediction rules, the use of MTBI populations other than hospital admissions, continued research on the effects of repeated

concussions, longer follow-up times with more measurement periods in longitudinal studies, an assessment of the differences between adults and

children, and an account for reverse causality and differential recall bias. Well-conducted studies in these areas will aid our understanding of

MTBI prognosis and assist clinicians in educating and treating their patients with MTBI.

Archives of Physical Medicine and Rehabilitation 2014;95(3 Suppl 2):S265-77

ª 2014 by the American Congress of Rehabilitation Medicine

Supported by the Ontario Neurotrauma Foundation (grant no. 2010-ABI-MTBIWHO-871). The funder was not involved in the design or preparation of the study protocol, or in the management of the

project, analysis or interpretation of data, or the preparation of the final article.

No commercial party having a direct financial interest in the results of the research supporting this article has conferred or will confer a benefit on the authors or on any organization with which the

authors are associated.

The findings and conclusions in this research are those of the authors alone and do not necessarily represent the official views or policies of the Centers for Disease Control and Prevention or any agency

of the United States government. Inclusion of individuals, programs, or organizations in this article does not constitute endorsement by the United States government.

0003-9993/14/$36 - see front matter ª 2014 by the American Congress of Rehabilitation Medicine

http://dx.doi.org/10.1016/j.apmr.2013.04.026

Delta:1_given name

Delta:1_surname

Delta:1_given name

Delta:1_surname

Delta:1_given name

Delta:1_surname

Delta:1_given name

Delta:1_surname

Delta:1_given name

Delta:1_surname

Delta:1_given name

Delta:1_surname

http://crossmark.crossref.org/dialog/?doi=10.1016/j.apmr.2013.04.026&domain=pdf


http://www.archives-pmr.org


S266 V.L. Kristman et al

In 2004, the World Health Organization (WHO) CollaboratingCentre for Neurotrauma Prevention, Management and Rehabili-tation Task Force on Mild Traumatic Brain Injury (WHO TaskForce) found the mild traumatic brain injury (MTBI) literature tobe large and of variable quality.1 The purpose of the InternationalCollaboration on MTBI Prognosis (ICoMP) was to update theWHO Task Force findings on prognosis and to comprehensivelysearch and critically review the quality of the new literature onMTBI prognosis. Overall, we accepted and based our findings on34% of the 299 eligible studies as having a low risk of bias.2 TheWHO Task Force accepted 28% of reviewed studies; thus, despitethe proliferation of MTBI research over the past decade, virtuallyno improvement was seen in study quality.1 These low acceptancerates indicate a literature plagued with poorly designed studiesand, as a result, many unanswered important clinical and researchquestions. The objectives of this article are to discuss issuesrelated to the definition and classification of MTBI; outlinecommon flaws including sources of bias, problems with studydesign, and issues with reporting; provide recommendations forimprovement; and identify priority areas of research onMTBI prognosis.

Definition of MTBI

The WHO Task Force report tabled 38 definitions from studiesincluded in their best evidence synthesis.1 Although many defini-tions had overlapping criteria, they also exhibited considerabledifferences and used varying terms for the condition includingMTBI, concussion, and minor head injury. Most definitions (62%)included Glasgow Coma Scale (GCS) scores as one or the onlycriterion, but not all applied the same GCS spectrum to defineMTBI. Others (38%) used varying criteria for loss of consciousness(LOC) and amnesia, or Abbreviated Injury Severity scores, or In-ternational Classification of Diseases (ICD) codes, or other criteria.Some definitions also allowed skull fractures or intracranial lesions.

Based on their findings, the WHO Task Force proposed anoperational definition derived from the definition developed by theMild Traumatic Brain Injury Committee of the Head Injury Inter-disciplinary Special Interest Group of the American Congress ofRehabilitationMedicine (ACRM).3 This definition reads as follows:

“MTBI is an acute brain injury resulting from mechanicalenergy to the head from external physical forces. Operationalcriteria for clinical identification include: (1) 1 or more of thefollowing: confusion or disorientation, LOC for 30 minutes or

List of abbreviations:

ACRM American Congress of Rehabilitation Medicine

ED emergency department

GCS Glasgow Coma Scale

ICD International Classification of Diseases

ICF International Classification of Functioning,

Disability and Health

ICoMP International Collaboration on MTBI Prognosis

LOC loss of consciousness

MTBI mild traumatic brain injury

PTA posttraumatic amnesia

RTW return to work

WHO World Health Organization

WHO Task Force WHO Collaborating Centre for Neurotrauma

Prevention, Management and Rehabilitation

Task Force on MTBI

less, posttraumatic amnesia for less than 24 hours, and/or othertransient neurologic abnormalities such as focal signs, seizure,and intracranial lesion not requiring surgery; (2) GCS score of13e15 after 30 minutes post-injury or later upon presentationfor health care. (3) These manifestations of MTBI must not bedue to drugs, alcohol, medications, caused by other injuries ortreatment for other injuries (eg, systemic injuries, facial in-juries, or intubation), caused by other problems (eg, psycho-logical trauma, language barrier, or coexisting medicalconditions), or caused by penetrating craniocerebralinjury.”1(p115)

This operational definition has similarities with the conceptualdefinition produced by a panel of experts from the U.S. Centers forDisease Control and Prevention’s MTBI Working Group in 2003.4

While the ACRM definition suggests that the GCS score of 13 to15 be assessed after 30 minutes postinjury, the WHO Task Forceproposed the use of a GCS score after 30 minutes or later onpresentation for health care, arguing that individuals with MTBIwill rarely be assessed at an emergency department (ED) within30 minutes. The proposed definition differed from the ACRMdefinition in 2 ways: (1) According to the ACRM criteria,disturbed mental status includes feeling “dazed,” but this was notincluded in the WHO Task Force definition; and (2) In the ACRMdefinition, focal neurologic deficits “may or may not be transient,”while the WHO Task Force definition states “other transientneurologic abnormalities.” Ruff et al5 correctly identify that “therational by the WHO Task Force for making these two changeswas not stated explicitly in their publication.”(p5) However, theextent to which the term “dazed” truly refers to disturbed mentalstatus could be debated.

Ruff5 also points out that neither definition provides guidelinesor specific recommendations for assessing the 4 key elements:LOC, posttraumatic amnesia (PTA), disorientation/confusion, andneurologic signs. This is a key issue, and we strongly recommendfurther study of the diagnostic accuracy of relevant measures.

The current review found a continuing lack of unity regardingMTBI definitions (table 1). While most studies include the GCSscore as 1 component, others apply only LOC or PTA-basedcriteria. Although most recent studies apply criteria that arecompatible with the ACRM and WHO Task Force criteria, inmany cases authors have selected a subset of patients with MTBIwho have a particular severity of injury, within the broader MTBIdefinition. The impact of different degrees of MTBI severity onthe prognosis is not known, and the variation of study samplecharacteristics hinders comparison of the results. We recommendfurther studies comparing the prognosis for subgroups with well-defined but different severity degrees. A more nuanced severitygrading based on GCS scores (GCS 15 vs 14 vs 13) or PTA levels,or both, may be promising. We need more information on thereliability of the GCS, especially in the 13 to 15 range.

Upper and lower severity delimitations of MTBI

The distinctions of MTBI from moderate TBI and from trivial headinjury are debated. Some authors have proposed that in patientswhere imaging demonstrates brain pathology, the diagnosis shouldbe complicated mild or moderate TBI.6 Since there is a higherfrequency of brain pathology in patients with a presenting GCSscore of 13, and the prognosis of these patients may be similar tothat of patients with moderate TBI, GCS scores of 13 may be betterclassified as being indicative of moderate TBI. However, given the

www.archives-pmr.org


Table 1 MTBI case definitions used in studies included in best-evidence synthesis of MTBI prognosis literature

MTBI Case Definition

Article

Citation*

GCS

GCS 13 or higher, absence of an injury-related intracranial abnormality on cranial CT scan, abnormality on neurologic

examination consistent with external trauma

1

GCS 13e14 2

GCS 13e14, witnessed LOC (time undefined), PTA, blunt head trauma 3

GCS 13e15, PTA <24h 4

GCS 13e14, blunt head trauma, LOC <30min, 2 acute symptoms of concussion as documented by ED medical personnel,

includes mild, complicated concussion

5e9

GCS 13e15 10e12

GCS 13e15; nonpenetrating, acute blunt head trauma 13

GCS 13e15, LOC <10min, PTA �24h, mild alteration in mental status (ie, dazed, disoriented, or confused immediately after

the injury), physiologic disruption of brain function

14

GCS (or children’s modified GCS) 13e15, LOC <15min, no neurosurgical intervention, and negative neuroimaging other

than linear skull fracture

15

GCS 13e15, LOC <15min, and at least 1 of the following symptoms on admission to hospital: headache, vertigo, amnesia,

nausea, vomiting

16

GCS 13e15, LOC <15min, PTA <60min, any alteration in mental state, lack of focal neurologic deficits 17

GCS 13e15, LOC <20min, PTA <24h 18

GCS 13e15, LOC �20min, normal CT or surgical findings; no history of prior head injury, alcoholism, drug intake, epilepsy,

cerebral disease, mental retardation or significant psychiatric disorder; willingness to participate in psychology tests

19

GCS 13e15, LOC <30min, closed head trauma as mechanism 20

GCS 13e15, LOC <30min and/or retrograde amnesia or PTA <24h and/or any alteration in mental state at time of injury;

state change caused by explosive event

21

GCS 13-15, LOC <30min, PTA and/or retrograde amnesia <24h, any alteration in mental state 22

GCS 13e15, LOC <30min, with or without PTA 23

GCS >13, LOC �30min due to trauma, PTA, 2 or more postconcussion symptoms 24

GCS 13e15, PTA between 3min and 4h 25

GCS 13e15, PTA (or transient confusion, disorientation, impaired consciousness) <24h, or signs of other neurologic or

neuropsychological dysfunction

26

GCS 13e15, LOC �30min, confusion or PTA (�24h) 27

GCS 13e15, minor closed head injury, brief LOC or amnesia 28

GCS 13e15, no LOC 29

GCS 13e15 (lowest in 24h), without neurologic or radiologic abnormalities. Admission to hospital 30

GCS 13e15 with or without CT abnormalities 31

GCS 13e15, no abnormality on CT/MRI scans, and no neurologic deficits 32e34

GCS 13e15, normal CT scan and neurologic examination findings, and absence of depressed skull fracture 35

GCS 14e15, blunt trauma to head, LOC <15min, PTA <1h, absence of focal neurologic signs 36

GCS 14e15 with LOC <30min and/or PTA <24h, injury within past 24h, blunt trauma to head 37

GCS 14e15, with or without abnormalities on head CT scans 38

GCS 15, blunt trauma, nonfocal neurologic examination findings, negative head CT scan. Use of modified GCS for young

children. CT scans performed on children with the following indications: LOC, amnesia, headache, seizure, nausea,

vomiting, confusion, sleepiness, or “unspecified symptoms.”

39

ICD codes

ICD codes 850, 800e803, 851e854 (ICD-8) and/or other fractures 40

ICD codes 850.99 (ICD-8) and S06.0 (ICD-10) 41

ICD codes 800, 801, 803, 850, 851e854 (ICD-8) 42, 43

ICD codes 850, 800e803, S06.0 (ICD-10) 44

ICD-9CM 45

ICD-9CM, AISZ2 46

ICD-9CM, LOC <1h, no traumatic intracranial lesions 47, 48

AIS

AIS 1e2 49

Alterations in consciousness (no GCS specified)ewithout GCS

LOCZundefined, or amnesia 50

LOCZ0, a few minutes of unconsciousness at most 51

LOC <1min, no immediate PTA 52

(continued on next page)

Methodological issues and research recommendations for prognosis after mild TBI S267



Table 1 (continued )

MTBI Case Definition

Article

Citation*

LOC >1min but <30min, no symptoms of depression and no self-reported neurologic history 53, 54

LOC <10min or no LOC, able to answer questions at time of admission 55

LOC <15min, admitted to ED 56

LOC <30min, PTA <24h, confusion or disorientation, other transient neurologic abnormalities (ie, focal signs, seizure,

intracranial lesions)

57

LOC or PTA <30min, no skull fracture 58

LOC <30min and/or PTA and/or disorientation and/or confusion <24h 59

LOC and/or PTA/retrograde amnesia, any noticeable change in mental status or consciousness, self-reported symptoms that

appeared after a collision on the field (eg, headache, dizziness, balance dysfunction, visual changes)

60

Concussion in Sport Group definition (1st International Conference, 2001) (complex pathophysiological process affecting

the brain, induced by traumatic biomechanical forces, rapid onset of short neurologic impairments and neuropathologic

changes, a graded set of clinical syndromes that may or may not involve LOC)

61e64

(1) Any observable alteration in mental status/level of consciousness such as LOC, retrograde amnesia, PTA, disorientation;

and/or (2) any self-reported symptoms after a collision such as “fogginess,” “grogginess,” headache, nausea/vomiting,

dizziness, balance problems, and/or visual changes

65

(1) Alteration of awareness or consciousness, including but not limited to being dinged, dazed, stunned, woozy, foggy,

amnesic, or less commonly, rendered unconsciousness or, even more rarely, experiencing seizure; and (2) signs and

symptoms commonly associated with postconcussion symptoms, including persistent headaches, vertigo,

lightheadedness, loss of balance, unsteadiness, syncope, near syncope, cognitive dysfunction, memory disturbance,

hearing loss, tinnitus, blurred vision, diplopia, visual loss, personality change, drowsiness, lethargy, fatigue, and

inability to perform usual daily activities

66e69

Other

AAN Concussion Grading Scale 70e75

AAN, Cantu, Guskiewicz, McCrea, Collins 76

American College/Congress of Rehabilitation Medicine definition 77e89

British Society of Rehabilitation Medicine classification 90

WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury 91

Acute brain injury resulting from mechanical energy to the head from external forces 92

Sudden acceleration/deceleration injury to the head resulting in temporary alteration in brain functions, LOC, blurred

vision, dizziness, amnesia, or memory impairment

93

An injury resulting from a blow to the head that caused an alteration in mental status and 1 or more of the following:

headache, nausea, vomiting, dizziness/balance problems, fatigue, difficulty sleeping, drowsiness, sensitivity to light or

noise, blurred vision, memory difficulty, difficulty concentrating

94

7-grade system for head injury 95

Change in mental status or LOC, PTA, retrograde amnesia, self-reported symptoms after collisiondheadache, balance

problems, dizziness, visual problems

96

None given 97

Poorly addressed 98

Impact to head, LOC <30min, PTA, disorientation/confusion, dizziness/unsteadiness, memory problems/forgetfulness,

concentration or attention problems

99, 100

Workplace Safety & Insurance BoardeOntario, Canada

Part of body: code 1100 (brain, head)

Nature of injury: code 6200 (concussion)

101

Abbreviations: AAN, American Academy of Neurology; AIS, Abbreviated Injury Scale; CT, computed tomography; ICD-9CM, International Classification of

Diseases, 9th Revision, Clinical Modification; MRI, magnetic resonance imaging.

* References

1. Rush BK, Malec JF, Moessner AM, Brown AW. Preinjury personality traits and the prediction of early neurobehavioral symptoms following mild

traumatic brain injury. Rehabil Psychol 2004;49:275-81.

2. De Silva MJ, for the CRASH Trial Collaborators. Patient outcome after traumatic brain injury in high-, middle- and low-income countries: analysis of

data on 8927 patients in 46 countries. Int J Epidemiol 2009;38:452-8.

3. Grubenhoff JA, Kirkwood MW, Deakyne S, Wathen J. Detailed concussion symptom analysis in a paediatric ED population. Brain Inj 2011;25:943-9.

4. Heitger MH, Jones RD, Dalrymple-Alford JC, Frampton CM, Ardagh MW, Anderson TJ. Motor deficits and recovery during the first year following mild

closed head injury. Brain Inj 2006;20:807-24.

5. Fay TB, Yeates KO, Taylor HG, et al. Cognitive reserve as a moderator of postconcussive symptoms in children with complicated and uncomplicated

mild traumatic brain injury. J Int Neuropsychol Soc 2010;16:94-105.

6. Taylor HG, Dietrich A, Nuss K, et al. Post-concussive symptoms in children with mild traumatic brain injury. Neuropsychology 2010;24:148-59.




7. Woodrome SE, Yeates KO, Taylor HG, et al. Coping strategies as a predictor of post-concussive symptoms in children with mild traumatic brain injury

versus mild orthopedic injury. J Int Neuropsychol Soc 2011;17:317-26.

8. Yeates KO, Taylor HG, Rusin J, et al. Longitudinal trajectories of postconcussive symptoms in children with mild traumatic brain injuries and their

relationship to acute clinical status. Pediatrics 2009;123:735-43.

9. Hajek CA, Yeates KO, Taylor HG, et al. Relationships among post-concussive symptoms and symptoms of PTSD in children following mild traumatic

brain injury. Brain Inj 2010;24:100-9.

10. de Leon MB, Kirsch NL, Maio RF, et al. Baseline predictors of fatigue 1 year after mild head injury. Arch Phys Med Rehabil 2009;90:956-65.

11. Iverson GL. Complicated vs uncomplicated mild traumatic brain injury: acute neuropsychological outcome. Brain Inj 2006;20:1335-44.

12. Pentland B, Hutton LS, Jones PA. Late mortality after head injury. J Neurol Neurosurg Psychiatry 2005;76:395-400.

13. Hollingworth W, Vavilala MS, Jarvik JG, et al. The use of repeated head computed tomography in pediatric blunt head trauma: factors predicting new

and worsening brain injury. Pediatr Crit Care Med 2007;8:348-56.

14. Brewer TL, Metzger BL, Therrien B. Trajectories of cognitive recovery following a minor brain injury. Res Nurs Health 2002;25:269-81.

15. Roncadin C, Guger S, Archibald J, Barnes M, Dennis M. Working memory after mild, moderate, or severe childhood closed head injury. Dev Neuro-

psychol 2004;25:21-36.

16. Lima DP, Simao Filho C, Abib Sde C, de Figueiredo LF. Quality of life and neuropsychological changes in mild head trauma. Late analysis and cor-

relation with S100B protein and cranial CT scan performed at hospital admission. Injury 2008;39:604-11.

17. Hou R, Moss-Morris R, Peveler R, Mogg K, Bradley BP, Belli A. When a minor head injury results in enduring symptoms: a prospective investigation of

risk factors for postconcussional syndrome after mild traumatic brain injury. J Neurol Neurosurg Psychiatry 2012;83:217-23.

18. Norrie J, Heitger M, Leathem J, Anderson T, Jones R, Flett R. Mild traumatic brain injury and fatigue: a prospective longitudinal study. Brain Inj

2010;24:1528-38.

19. Shanmukhi S, Panigrahi M. Serial pattern learning after mild head injury. Neurol India 2003;51:518-21.

20. Levin HS, Hanten G, Roberson G, et al. Prediction of cognitive sequelae based on abnormal computed tomography findings in children following mild

traumatic brain injury. J Neurosurg Pediatr 2008;1:461-70.

21. Kennedy JE, Cullen MA, Amador RR, Huey JC, Leal FO. Symptoms in military service members after blast mTBI with and without associated injuries.

NeuroRehabilitation 2010;26:191-7.

22. Cooper DB, Mercado-Couch JM, Critchfield E, et al. Factors influencing cognitive functioning following mild traumatic brain injury in OIF/OEF burn

patients. NeuroRehabilitation 2010;26:233-8.

23. Stulemeijer M, van der Werf S, Borm GF, Vos PE. Early prediction of favourable recovery 6 months after mild traumatic brain injury. J Neurol Neurosurg

Psychiatry 2008;79:936-42.

24. McLean SA, Kirsch NL, Tan-Schriner CU, et al. Health status, not head injury, predicts concussion symptoms after minor injury. Am J Emerg Med

2009;27:182-90.

25. Heitger MH, Macaskill MR, Jones RD, Anderson TJ. The impact of mild closed head injury on involuntary saccadic adaptation: evidence for the

preservation of implicit motor learning. Brain Inj 2005;19:109-17.

26. Tham SW, Palermo TM, Vavilala MS, et al. The longitudinal course, risk factors, and impact of sleep disturbances in children with traumatic brain

injury. J Neurotrauma 2012;29:154-61.

27. Bell KR, Hoffman JM, Temkin NR, et al. The effect of telephone counselling on reducing post-traumatic symptoms after mild traumatic brain injury: a

randomised trial. J Neurol Neurosurg Psychiatry 2008;79:1275-81.

28. Halley MK, Silva PD, Foley J, Rodarte A. Loss of consciousness: when to perform computed tomography? Pediatr Crit Care Med 2004;5:230-3.

29. Gil S, Caspi Y, Ben-Ari IZ, Koren D, Klein E. Does memory of a traumatic event increase the risk for posttraumatic stress disorder in patients with

traumatic brain injury? A prospective study. Am J Psychiatry 2005;162:963-9.

30. Anderson V, Le Brocque R, Iselin G, et al. Adaptive ability, behavior and quality of life pre and posttraumatic brain injury in childhood. Disabil Rehabil

2012;34:1639-47.

31. Doctor JN, Castro J, Temkin NR, Fraser RT, Machamer JE, Dikmen SS. Workers’ risk of unemployment after traumatic brain injury: a normed comparison.

J Int Neuropsychol Soc 2005;11:747-52.

32. Anderson V, Catroppa C, Haritou F, Morse S, Rosenfeld J. Identifying factors contributing to child and family outcome 30 months after traumatic brain

injury in children. J Neurol Neurosurg Psychiatry 2005;76:401-8.

33. Anderson V, Catroppa C, Morse S, Haritou F, Rosenfeld J. Functional plasticity or vulnerability after early brain injury? Pediatrics 2005;116:1374-82.

34. Kenardy J, Le Brocque R, Hendrikz J, Iselin G, Anderson V, McKinlay L. Impact of posttraumatic stress disorder and injury severity on recovery in

children with traumatic brain injury. J Clin Child Adolesc Psychol 2012;41:5-14.

35. Luis CA, Mittenberg W. Mood and anxiety disorders following pediatric traumatic brain injury: a prospective study. J Clin Exp Neuropsychol

2002;24:270-9.

36. de Kruijk JR, Leffers P, Meerhoff S, Rutten J, Twijnstra A. Effectiveness of bed rest after mild traumatic brain injury: a randomised trial of no versus six

days of bed rest. J Neurol Neurosurg Soc 2002;73:167-72.

37. Nygren-de Boussard C, Lundin A, Karlstedt D, Edman G, Bartfai A, Borg J. S100 and cognitive impairment after mild traumatic brain injury. J Rehabil

Med 2005;37:53-7.

38. Blinman TA, Houseknecht E, Snyder C, Wiebe DJ, Nance ML. Postconcussive symptoms in hospitalized pediatric patients after mild traumatic brain

injury. J Pediatr Surg 2009;44:1223-8.

39. Spencer MT, Baron BJ, Sinert R, Mahmoud G, Punzalan C, Tintinalli A. Necessity of hospital admission for pediatric minor head injury. Am J Emerg Med

2003;21:111-4.

40. Nielsen AS, Mortensen PB, O’Callaghan E, Mors O, Ewald H. Is head injury a risk factor for schizophrenia? Schizophr Res 2002;55:93-8.

41. Christensen J, Pedersen MG, Pedersen CB, Sidenius P, Olsen J, Vestergaard M. Long-term risk of epilepsy after traumatic brain injury in children and

young adults: a population-based cohort study. Lancet 2009;373:1105-10.

42. Teasdale TW, Engberg AW. Suicide after traumatic brain injury: a population study. J Neurol Neurosurg Psychiatry 2001;71:436-40.

43. Teasdale TW, Engberg AW. Cognitive dysfunction in young men following head injury in childhood and adolescence: a population study. J Neurol

Neurosurg Psychiatry 2003;74:933-6.

44. Rugbjerg K, Ritz B, Korbo L, Martinussen N, Olsen JH. Risk of Parkinson’s disease after hospital contact for head injury: population based case-control

study. BMJ 2008;337:a2494.




45. Fann JR, Burington B, Leonetti A, Jaffe K, Katon WJ, Thompson RS. Psychiatric illness following traumatic brain injury in an adult health maintenance

organization population. Arch Gen Psychiatry 2004;61:53-61.

46. Selassie AW, Fakhry SM, Ford DW. Population-based study of the risk of in-hospital death after traumatic brain injury: the role of sepsis. J Trauma

2011;71:1226-34.

47. Massagli TL, Fann JR, Burington BE, Jaffe KM, Katon WJ, Thompson RS. Psychiatric illness after mild traumatic brain injury in children. Arch Phys Med

Rehabil 2004;85:1428-34.

48. Rockhill CM, Fann JR, Fan MY, Hollingworth W, Katon WJ. Healthcare costs associated with mild traumatic brain injury and psychological distress in

children and adolescents. Brain Inj 2010;24:1051-60.

49. Babikian T, Satz P, Zaucha K, Light R, Lewis RS, Asarnow RF. The UCLA longitudinal study of neurocognitive outcomes following mild pediatric

traumatic brain injury. J Int Neuropsychol Soc 2011;17:886-95.

50. Goldman SM, Tanner CM, Oakes D, Bhudhikanok GS, Gupta A, Langston JW. Head injury and Parkinson’s disease risk in twins. Ann Neurol 2006;60:65-72.

51. Helmes E, Ostbye T, Steenhuis RE. Incremental contribution of reported previous head injury to the prediction of diagnosis and cognitive functioning

in older adults. Brain Inj 2011;25:338-47.

52. Da Dalt L, Andreola B, Facchin P, Gregolin M, Vianello A, Battistella PA. Characteristics of children with vomiting after minor head trauma: a case-

control study. J Pediatr 2007;150:274-8.

53. Suhr JA, Gunstad J. “Diagnosis threat”: the effect of negative expectations on cognitive performance in head injury. J Clin Exp Neuropsychol

2002;24:448-57.

54. Suhr JA, Gunstad J. Further exploration of the effect of “diagnosis threat” on cognitive performance in individuals with mild head injury. J Int

Neuropsychol Soc 2005;11:23-9.

55. Korinthenberg R, Schreck J, Weser J, Lehmkuhl G. Post-traumatic syndrome after minor head injury cannot be predicted by neurological in-

vestigations. Brain Dev 2004;26:113-7.

56. Stovner LJ, Schrader H, Mickeviciene D, Surkiene D, Sand T. Headache after concussion. Eur J Neurol 2009;16:112-20.

57. Gardner A, Shores EA, Batchelor J. Reduced processing speed in rugby union players reporting three or more previous concussions. Arch Clin Neu-

ropsychol 2010;25:174-81.

58. Bower JH, Maraganore DM, Peterson BJ, McDonnell SK, Ahlskog JE, Rocca WA. Head trauma preceding PD: a case-control study. Neurology

2003;60:1610-5.

59. Ozen LJ, Fernandes MA. Effects of “diagnosis threat” on cognitive and affective functioning long after mild head injury. J Int Neuropsychol Soc

2011;17:219-29.

60. Lau BC, Collins MW, Lovell MR. Sensitivity and specificity of subacute computerized neurocognitive testing and symptom evaluation in predicting

outcomes after sports-related concussion. Am J Sports Med 2011;39:1209-16.

61. Collie A, Makdissi M, Maruff P, Bennell K, McCrory P. Cognition in the days following concussion: comparison of symptomatic versus asymptomatic

athletes. J Neurol Neurosurg Psychiatry 2006;77:241-5.

62. Covassin T, Elbin RJ, Nakayama Y. Tracking neurocognitive performance following concussion in high school athletes. Phys Sportsmed 2010;38:87-93.

63. Makdissi M, Darby D, Maruff P, Ugoni A, Brukner P, McCrory PR. Natural history of concussion in sport: markers of severity and implications for

management. Am J Sports Med 2010;38:464-71.

64. Makdissi M, McCrory P, Ugoni A, Darby D, Brukner P. A prospective study of postconcussive outcomes after return to play in Australian football. Am J

Sports Med 2009;37:877-83.

65. Lovell MR, Collins MW, Iverson GL, et al. Recovery from mild concussion in high school athletes. J Neurosurg 2003;98:296-301.

66. Pellman EJ, Lovell MR, Viano DC, Casson IR. Concussion in professional football: recovery of NFL and high school athletes assessed by computerized

neuropsychological testingdpart 12. Neurosurgery 2006;58:263-74.

67. Pellman EJ, Viano DC, Casson IR, Arfken C, Feuer H. Concussion in professional football: players returning to the same gamedpart 7. Neurosurgery

2005;56:79-92.

68. Pellman EJ, Viano DC, Casson IR, Arfken C, Powell J. Concussion in professional football: injuries involving 7 or more days outdpart 5. Neurosurgery

2004;55:1100-19.

69. Pellman EJ, Viano DC, Casson IR, et al. Concussion in professional football: repeat injuriesdpart 4. Neurosurgery 2004;55:860-76.

70. Bruce JM, Echemendia RJ. Concussion history predicts self-reported symptoms before and following a concussive event. Neurology 2004;63:1516-8.

71. Covassin T, Schatz P, Swanik CB. Sex differences in neuropsychological function and post-concussion symptoms of concussed collegiate athletes.

Neurosurgery 2007;61:345-51.

72. Covassin T, Stearne D, Elbin R III. Concussion history and postconcussion neurocognitive performance and symptoms in collegiate athletes. J Athl

Train 2008;43:119-24.

73. Fazio VC, Lovell MR, Pardini JE, Collins MW. The relation between post concussion symptoms and neurocognitive performance in concussed athletes.

NeuroRehabilitation 2007;22:207-16.

74. Field M, Collins MW, Lovell MR, Maroon J. Does age play a role in recovery from sports-related concussion? A comparison of high school and collegiate

athletes. J Pediatr 2003;142:546-53.

75. Iverson GL, Lovell MR, Collins MW. Interpreting change on ImPACT following sport concussion. Clin Neuropsychol 2003;17:460-7.

76. Erlanger D, Kaushik T, Cantu R, et al. Symptom-based assessment of the severity of a concussion. J Neurosurg 2003;98:477-84.

77. Davis CH. Self-perception in mild traumatic brain injury. Am J Phys Med Rehabil 2002;81:609-18.

78. Dischinger PC, Ryb GE, Kufera JA, Auman KM. Early predictors of postconcussive syndrome in a population of trauma patients with mild traumatic

brain injury. J Trauma 2009;66:289-97.

79. Friedland JF, Dawson DR. Function after motor vehicle accidents: a prospective study of mild head injury and posttraumatic stress. J Nerv Ment Dis

2001;189:426-34.

80. Gagnon I, Swaine B, Friedman D, Forget R. Children show decreased dynamic balance after mild traumatic brain injury. Arch Phys Med Rehabil

2004;85:444-52.

81. Gagnon I, Swaine B, Friedman D, Forget R. Visuomotor response time in children with a mild traumatic brain injury. J Head Trauma Rehabil

2004;19:391-404.

82. Gagnon I, Swaine B, Friedman D, Forget R. Exploring children’s self-efficacy related to physical activity performance after a mild traumatic brain

injury. J Head Trauma Rehabil 2005;20:436-49.




83. Kashluba S, Paniak C, Blake T, Reynolds S, Toller-Lobe G, Nagy J. A longitudinal, controlled study of patient complaints following treated mild

traumatic brain injury. Arch Clin Neuropsychol 2004;19:805-16.

84. Landre N, Poppe CJ, Davis N, Schmaus B, Hobbs SE. Cognitive functioning and postconcussive symptoms in trauma patients with and without mild

TBI. Arch Clin Neuropsychol 2006;21:255-73.

85. Preiss-Farzanegan SJ, Chapman B, Wong TM, Wu J, Bazarian JJ. The relationship between gender and postconcussion symptoms after sport-related

mild traumatic brain injury. PM R 2009;1:245-53.

86. Sundstrom A, Marklund P, Nilsson LG, et al. APOE influences on neuropsychological function after mild head injury. Neurology 2004;62:1963-6.

87. Sundstrom A, Nilsson LG, Cruts M, Adolfsson R, Van Broeckhoven C, Nyberg L. Fatigue before and after mild traumatic brain injury: pre-post-injury

comparisons in relation to apolipoprotein E. Brain Inj 2007;21:1049-54.

88. Tellier A, Marshall SC, Wilson KG, Smith A, Perugini M, Stiell IG. The heterogeneity of mild traumatic brain injury: where do we stand? Brain Inj

2009;23:879-87.

89. Cooper DB, Kennedy JE, Cullen MA, Critchfield E, Amador RR, Bowles AO. Association between combat stress and post-concussive symptom reporting

in OEF/OIF service members with mild traumatic brain injuries. Brain Inj 2011;25:1-7.

90. Hawley CA, Ward AB, Magnay AR, Long J. Outcomes following childhood head injury: a population study. J Neurol Neurosurg Psychiatry 2004;75:737-42.

91. Lange RT, Iverson GL, Rose A. Post-concussion symptom reporting and the “good-old-days” bias following mild traumatic brain injury. Arch Clin

Neuropsychol 2010;25:442-50.

92. Kristman VL, Cote P, Hogg-Johnson S, et al. The burden of work disability associated with mild traumatic brain injury in Ontario compensated workers:

a prospective cohort study. Open Occup Health Safety J 2010;2:1-8.

93. Guskiewicz KM, Ross SE, Marshall SW. Postural stability and neuropsychological deficits after concussion in collegiate athletes. J Athl Train

2001;36:263-73.

94. Guskiewicz KM, McCrea M, Marshall SW, et al. Cumulative effects associated with recurrent concussion in collegiate football players: the NCAA

concussion study. JAMA 2003;290:2549-55.

95. AbdelMalik P, Husted J, Chow EW, Bassett AS. Childhood head injury and expression of schizophrenia in multiply affected families. Arch Gen Psy-

chiatry 2003;60:231-6.

96. Lau BC, Kontos AP, Collins MW, Mucha A, Lovell MR. Which on-field signs/symptoms predict protracted recovery from sport-related concussion among

high school football players? Am J Sports Med 2011;39:2311-8.

97. Pellman EJ, Lovell MR, Viano DC, Casson IR, Tucker AM. Concussion in professional football: neuropsychological testingdpart 6. Neurosurgery

2004;55:1290-305.

98. Bruce JM, Echemendia RJ. History of multiple self-reported concussions is not associated with reduced cognitive abilities. Neurosurgery 2009;64:100-6.

99. Cassidy JD, Boyle E, Carroll LJ. Population-based, inception cohort study of the incidence, course, and prognosis of mild traumatic brain injury after

motor vehicle collisions. Arch Phys Med Rehabil 2014;95(3 Suppl 2):S278-85.

100. Hartvigsen J, Boyle E, Cassidy JD, Carroll LJ. Mild traumatic brain injury after motor vehicle collisions: what are the symptoms and who treats them? A

population-based 1-year inception cohort study. Arch Phys Med Rehabil 2014;95(3 Suppl 2):S286-94.

101. Kristman VL, Cote P, Yang X, Hogg-Johnson S, Vidmar M, Rezai M. Health care utilization of workers’ compensation claimants associated with mild

traumatic brain injury: a historical population-based cohort study of workers injured in 1997-1998. Arch Phys Med Rehabil 2014;95(3 Suppl 2):S295-302.


current lack of validity and reliability information on the GCS,keeping the current GCS score distinction between mild andmoderate TBI seems justified. Further, new imaging methods, suchas magnetic resonance imaging, may reveal more traumatic pa-thology than current methods, so it seems reasonable to restrictcriteria for MTBI to clinical findings (signs and symptoms ofdisturbed neurologic functions identified by a structured assess-ment) in the acute phase or at earliest convenience.

The delineation of MTBI from trivial head injury is morechallenging. The central clinical features of MTBI (LOC, PTA,disorientation/confusion, neurologic signs) need to be betterdefined, and more research is needed regarding optimal methodsfor assessing these features.5 In the meantime, studies shouldexplicitly state how these features have been assessed.

PTA is considered by many to be the most specific clinicaldiagnostic indicator of brain injury. Even this is not straightfor-ward since nonbrain injuries may cause pain or stress, which hasthe potential to cause mild degrees of memory and other cognitivedysfunction. There are also inherent difficulties with the termPTA: in clinical practice, the term PTA commonly refers to a stateencompassing not only amnesia but also confusion and disorien-tation.7 It is redundant and possibly misleading to list confusionand disorientation separately from PTA in MTBI definitions. Thealternative term posttraumatic confusion is sometimes used todescribe the same clinical entity.8

Further, although focal neurologic symptoms and signs may bepresent, they are not common findings post-MTBI and not typicalindicators of MTBI. Therefore, these may be considered compatiblewith MTBI but not part of the definition, which then would be based


solely on markers of “consciousness” (ie, wakefulness and amnesia).A minimal requirement for diagnosis of MTBI would then beamnesia for the moment of the injury. Penetrating injuries, such asgunshot wounds, are typically focal and may occur with unaffected“consciousness” and thus do not fit into the MTBI concept.

Another issue relates to exposures to external mechanicalforces other than those transferred by a blow to the head. Menonet al9 comment that “this understanding that both milder insultsand less typical presentations now fit under the TBI diagnosticumbrella has increased the overlap with (and diagnostic confoundby) non-TBI pathologies. These overlapping conditions can resultin diagnostic ambiguity, which may confound precise epidemio-logic description, appropriate clinical management, and rationalresearch strategy development.”9(p1637) We agree but recommendthat the use of the term MTBI be restricted to injuries caused bydirect blunt head trauma (ie, when the head hits or is hit by anobject). Then, other exposures (eg, blasts, whiplash) are excludedfrom the MTBI definition. This does not mean such exposuresmay not impact brain function, but until there is more clarity, itmakes sense not to cloud the MTBI concept. Obviously, pressurewaves of blast injuries may cause a person to fall or objects tomove and hit the skull and indirectly cause an MTBI.

Terminology

MTBI and concussion are currently used in parallel and withoverlapping definitions. This hampers communication and com-parison of study results. The terms differ semantically and



perceptually by many clinicians and laymen.10 Inherent in theconcept of concussion is the acute occurrence of symptoms andsigns of disturbed consciousness that are transient and correspondto temporary dysfunction rather than permanent loss of brain tissueand function. In contrast, MTBI is conceptually associated with amechanically induced lesion of brain tissue. One approach inaccordance with most clinicians and laymen would be to restrict theterm concussion to the mildest forms of MTBI. This may still beproblematic and confusing. Keeping both “concussion” and“MTBI” as synonyms requires translation for MTBI to ICD codes;further efforts here are strongly recommended.

Some admissible studies used ICD codes (see table 1). Thediagnostic accuracy of these codes depends not only on the codes,but also on the knowledge and skills of the coder. The accuracy ofcoding MTBI in an emergency setting is likely low because ofmany factors, including the lack of a universal definition. Moreconspicuous injuries may risk missing a diagnosis of MTBI. Puljulaet al11 demonstrated that an MTBI diagnosis may be missed inpatients who present with craniofacial fractures. A recent system-atic review12 of ICD-10 codes used for neurotrauma surveillanceillustrates a wide variation of coding for TBI and MTBI. The au-thors discuss the diagnostic accuracy of using other injury codes, inparticular those for facial injuries or unspecified head injuries, asindicators of TBI. ICD-11 is currently under development, withplans for a final version in 2015.13 This is an opportunity toestablish consensus in the field. The ICoMP recommends, insteadof proposing yet another MTBI definition, an international effort tostandardize and refine existing definitions, something best done viacollaboration with the WHO in developing ICD-11. Input fromclinicians and researchers will ensure that ICD-11 codes for MTBIare as valid and reliable as possible. It is concerning that several odddiagnoses, not compatible with the current understanding of MTBIand existing definitions, are present at this relatively late stage in theICD-11 development process.13

Some diagnostic criteria rely on particular clinical and radio-logic examinations, at a specific point in time. If a GCS score is anessential part of diagnostic criteria, it follows that patients whowere not assessed with the GCS cannot receive that diagnosis,even if other clinical information is indicative. Similarly, patientswith identical injuries and an identical postacute clinical coursemay receive different diagnostic codes if different radiologic in-vestigations have been performed. For example, a patient withMTBI may receive a concussion diagnosis if computed tomog-raphy of the brain is performed and the findings are normal, butmay receive a diffuse brain injury diagnosis if magnetic resonanceimaging of the brain is performed and shows diffuse axonal injury(which may not be visible on computed tomography of the brain).Importantly, we do not suggest magnetic resonance imagingscanning for all patients with concussion, but the example isillustrative of the difficulties accompanying increasing but vari-able access to radiologic investigation. For research it may benecessary to restrict inclusion of patients to those assessed withparticular clinical or radiologic methods, but clinically such dis-parities need to be acknowledged and addressed in future clinicalguidance and in the development of ICD-11.

In summary, there is an obvious need for continued efforts toreach global consensus on a common terminology and definitionof MTBI. The key to achieving this is continued research on thepsychometric properties of key components and instruments usedto define MTBI, such as GCS, PTA, and radiologic results. Un-derstanding the diagnostic accuracy and measurement validity for

particular instruments will aid in minimizing the risk formisclassification. Appreciating the reliability of an instrument willreflect the stability and usefulness for defining MTBI. Recog-nizing the responsiveness of an instrument will clarify its ability todetect changes over time. ICoMP recommends further study onthe definition and measurement of MTBI.

Quality of Existing Research

We found important biases and methodological flaws in the designand conduct of MTBI prognosis studies (table 2). The mostcommon reasons for nonacceptance or exclusion in the ICoMPreview are highlighted in this section.

Bias

Lack of control for confoundingConfounding is a fundamental bias in health research, and lack ofinformation on, or consideration of, confounding factors wherewarranted was the primary reason for not accepting studies (seetable 2). Confounding occurs when the association between anexposure and an outcome is investigated, but the exposure andoutcome are strongly associated with a third variable. Lack ofcontrol of confounding variables, either by study design or byanalysis, may result in biased findings. In the case of prognosticfactors, we cannot assume that 1 factor is associated with recoveryunless we account for confounders of that association. Researchersneed to a priori consider the prognostic factors strongly associatedwith the outcome of interest and include these in their analyses todetermine independent effects and control for them if necessary.

For example, a prospective cohort study is conducted to deter-mine the effect of MTBI diagnosed in the ED on self-reportedsymptom recovery 5 days postinjury. Patients are compared with anunexposed (ie, no MTBI) group of persons visiting the EDfor musculoskeletal injury. Individuals with other symptoms(eg, depression) or preexisting health conditions (eg, diabetes) maybe less likely to report recovery 5 days postinjury. If more people inthe MTBI group have these other symptoms or health conditionsthan in the musculoskeletal group, we may conclude that those inthe MTBI group reported less recovery at 5 days postinjury thanthose with the musculoskeletal injuries. However, because we failedto consider the effects of other symptoms or preexisting healthconditions, we have erroneously identified the MTBI as the factorassociated with poor recovery, when in fact, the association betweenMTBI and recovery may be completely explained by other symp-toms or preexisting conditions (fig 1). Our review of adult prognosisof MTBI includes several accepted articles that used an orthopedicinjury control group to compare subjective outcomes after MTBI.14

Although the use of orthopedic injury controls is a good approach,one must consider factors that could confound and potentiallyoverestimate an association between MTBI and poor outcome.

Information biasInformation bias occurs when there is a flaw in measurement ofthe MTBI, outcome, or prognostic or confounding factors thatresults in varying accuracy of information between groups understudy. As mentioned in the definition section of this paper, mea-sures must be valid, reliable, and responsive.

Iverson and Gaetz15 discuss psychometric issues with usingvarious measures to assess recovery. Differential measurement of



Table 2 Methodological issues associated with articles deemed

inadmissible from the MTBI prognosis best-evidence synthesis

Methodological Issue

No. of

Studies*

GCS criteria

GCS 13 not included 2

MTBI related

Few/no MTBI-specific data 4

Limited analysis of MTBI patients 2

Poor definition of MTBI 38

Study populations/sample selection

Lack of characteristics of study population 17

Lack of information about sample selection

methods

15

Lack of true control group 15

Poor group selection 17

Poor randomization 5

Small sample size 32

(Potential) biases/reliability/validity

Attrition bias 4

Information bias 18

Interviewer bias 2

Low/no external validity/generalizability 8

Poor information on/consideration for

confounding factors

56

Recall bias 9

Reverse causality 2

Selection bias 31

Methods

Incomplete data 4

Nonblinded study 12

Poor study design 4

Unclear reporting of methods, ascertainment

of exposure, ascertainment of outcome

42

Analysis issues

Substandard or inappropriate analysis strategies 20

No comparison of experimental groups 4

No description of analysis 3

Retrospective/cross-sectional analysis 8

Follow-up

No indication of completeness of follow-up 9

Poor or lack of follow-up 14

Results

No baseline measures 3

Poor outcome data 10

Poor reporting of results 17

Participation

Compliance not fully explained 1

Low participation rate/high dropout 7

No dropout analysis 2

Unknown participation rate 7

Conclusions

Methodology and conclusions mismatch 2

Others

Hardly possible to read 1

Article has little to do with MTBI prognosis 2

Focus on PCS 1

Abbreviation: PCS, postconcussion syndrome.

* This columndoesnot add to the total number of inadmissible studies, as

many articles hadmore than 1 reason associated with their inadmissibility.



the outcome is a concern. For example, parents of children withMTBI will likely be more alert to symptoms or behavior problemsin their children than parents of uninjured children. This mayoverestimate the effect of the MTBI on symptoms reported,entirely because of observation or recall bias.

Selection and attrition biasSelection bias is a commonly overlooked problem in prognosticstudies of MTBI.16 It occurs when systematic differences occurbetween those who participate in a study and those who do not, andthose differences are related to the prognostic factor and outcomeof interest. Given the difficulty in assessing prognostic factors andoutcomes in nonparticipants, it is vital to obtain the highestparticipation rates possible. Investigators need to ensure that theydo not create a selected sample through overzealous inclusion andexclusion criteria or allow opportunities for self-selection intostudies. For example, Luoto et al16 demonstrated how the use oftypical inclusion/exclusion criteria could result in enrolling only3.4% of the eligible patients with MTBI. Regardless of theparticipation rate, clear descriptions of how samples were obtainedare necessary to evaluate the likelihood of selection bias.

Selection bias also occurs when the time frame for MTBI is notpart of the inclusion/exclusion criteria. All outcomes will have alatent time associated with them (ie, time for the outcome todevelop). For example, headache generally occurs fairly soon afteran MTBI. If the MTBI group under study consists of patientsrecruited from 2 days to 2 years after the injury, it is likely that onlythe most recently injured will still have headache symptoms thatcould be attributed to the MTBI. Therefore, selection of participantswith MTBI that includes those with recent and older injuries willresult in biased findings because of the lack of homogeneity in theMTBI group (also called zero-time bias because the time from theMTBI to inclusion in the study differs among cases). Similarly, if theoutcome is dementia, the latent period may be many years. Inclusionof mainly recent MTBIs would underestimate the association be-tween MTBI and dementia if the MTBI group has not had the op-portunity to develop the outcome similarly to the comparison group.

For longitudinal designs, it is imperative that authors reportattrition rates throughout the follow-up period. Otherwise it isimpossible to assess the influence of selective participation on thefindings. Additionally, we recommend that authors make consid-erable effort to understand the reasons behind attrition and toconduct sensitivity analyses to assess the potential impact on thestudy’s findings. At the very least, those who drop out of studiesneed to be compared with those who do not.

Problems with design

Lack of a true comparison groupAppropriate comparison groups are needed to determine whetherfactors associated with MTBI prognosis are specific to MTBI.Comparison groups must arise from the same source population asthe participants exposed to MTBI so that if a comparison subjectwere to sustain an MTBI, that individual would be captured in theMTBI group. Consider a cohort study to determine whether MTBIpatients identified in the ED are more likely to develop depressionthan a group of randomly selected patients without MTBI visitingthe same ED. The non-MTBI comparator group gives a referencepoint for comparison and reflects the “base risk” of depression inthose not exposed to MTBI. If the comparison group is not drawnfrom the same population as the patients with MTBI, it will notadequately represent the distribution of depression to which the


Fig 1 Example of confounding in MTBI research. This diagram

shows that if cases of MTBI are more likely than a comparison group of

musculoskeletal (MSK) injuries to be depressed, and those with

depression are more likely to report poor recovery 5 days postinjury,

then the MTBI may erroneously be considered the cause of poor re-

covery. This illustrates a confounding effect due to depression.


patients with MTBI should be compared. Consequently, case re-ports and case series, such as those described by McKee et al,17

are not appropriate for determining causal outcomes after MTBIbecause they have no control group by design. Hence, these de-signs were not included in the ICoMP review.

Use of small samplesSmall samples were used in 32 of the articles not accepted in theICoMP review (see table 2). Using small samples may lead toinadequate statistical power, meaning the study will be unable todemonstrate a statistically significant association even if one trulyexists. Sample size considerations need to be contemplated apriori, during the design phase, to ensure that clinically importantdifferences can be detected.

Prevalence versus incidenceTemporality between the MTBI exposure and outcome (eg, returnto work [RTW]) is important. Cross-sectional studies, which mea-sure prevalence, assess exposure and outcome at the same time (eg,assessing concussion symptoms and work status in a 1-time survey)and do not provide evidence for prognosis. Historical and pro-spective cohort studies, which measure incidence, assess theexposure before the outcome (eg, assessing concussion symptomsin the occupational physician’s office or through medical records,and then assessing subsequent RTW through occupational records)and provide evidence for prognosis. The ICoMP accepted studieswhere the exposure and outcome were obtained at the same time aslong as the exposure was a historical assessment (ie, history ofMTBI).18 It can be challenging to differentiate these designs;therefore, researchers must make it clear in their manuscript exactlyhow the exposure and outcome are assessed.

General problems

Unclear reporting of source population, methods, and resultsMany studies were not accepted because of the lack of informationprovided (see table 2). Source populations, samples, and com-parison groups were often poorly described. Generalizabilitycannot be determined without information on the source popula-tion. Selection bias cannot be assessed when the study groups arenot adequately described. Methods used to ascertain MTBI andpost-MTBI outcomes are crucial to confirming a lack of bias in astudy and must be described. Other articles excluded importantfindings such as response rates and numbers of participants.

Authors of MTBI prognosis manuscripts should follow theSTROBE (Strengthening the Reporting of Observational Studiesin Epidemiology) guidelines for reporting observational studies19

and the CONSORT (Consolidated Standards of Reporting Trials)guidelines for reporting randomized controlled trials.20

Inappropriate data analysis techniquesMany of the analytic limitations associated with MTBI prognosisstudies are connected to the issue of confounding. As highlightedearlier, some studies entirely lack consideration of confoundingwhen warranted. Those using multivariable modeling to accountfor confounding often try to include too many variables in themodel without considering the available sample size. This leads tolarge standard errors, large confidence intervals, and in extremesituations an inability to run the model.21

Methodological Issues and Research Priorities

Methodological issues

Even among accepted studies, nonfatal methodological issuesexisted. Table 3 summarizes some of the main issues and high-lights practical recommendations to assist clinical researchers inaddressing them.

Outcome measuresTo date, all systematic reviews on outcome measures after TBIfocus on moderate to severe injury22 or mix results from all braininjury severities together.23 No systematic review has focusedexclusively on describing validated outcome measures after MTBI.Hence, it is the authors’ responsibility to demonstrate the mea-surement properties (validity, reliability, responsiveness) of theoutcome measures used in their studies. Unfortunately, very few ofthe articles in the ICoMP reviews did this.

A research priority is to determine a set of clinically meaningfuloutcome measures specific to MTBI. Some authors have recom-mended the Glasgow Outcome ScaleeExtended to assess globaloutcome,24 theRivermeadPostconcussionSymptomsQuestionnaireto assess postconcussion symptomatology, the Medical OutcomesStudy 36-Item Short Form Health Survey for quality of life, andRTWas a social and economic outcome.22 ICoMPhighlights as a gapin the literature the identification and measurement assessment ofMTBI outcome and recommends the assessment (or development)of outcome measures in conjunction with the components of theInternational Classification of Functioning, Disability and Health(ICF).25 The ICF is a universal reference framework for functioningand health and is based on a comprehensive biopsychosocial modelof health. Additionally, newmeasures should comply with standardsset by COSMIN (COnsensus-based Standards for the selection ofhealth status Measurement Instruments).26

Follow-upFew accepted studies followed up people with MTBI usingreasonable intervals from the initial postinjury phase to 12 monthspostinjury. We recommend studies conduct longer follow-up pe-riods and include more measurement points throughout thefollow-up. A reasonable follow-up period depends on the outcomebeing assessed, but most persons with MTBI recover within weeksto months of their injury, so follow-up periods need to account forthis. Future research should determine ideal follow-up times forspecific outcomes.



Table 3 Practical checklist to aid clinical researchers designing and conducting studies of prognosis after MTBI

Area of

Difficulty Recommendations

Examples of Possible Measures to Comply With

Recommendations

MTBI

definition

Clearly defined definition, compatible with WHO Task Force

2004, including:

Mechanism of injury Documentation, specification of source of information

(eg, patient report, ambulance journal, eyewitness)

Severity within MTBIdupper and lower limits

specific to the study in question, in terms of:a. Acute disturbance of consciousness

b. Duration of PTA (a concept encompassing both amnesia

and confusion/disorientation)

a. Assessment of GCS at defined time points after injury,

witness reports, operational definitions of potentially

vague terms such as “dazed,” “seeing stars”

b. Use of well-characterized PTA scalesdeg, Westmead PTA

Scale, Modified Oxford PTA Scale, Nijmegen PTA Scale

Time after

injury

Homogenous samples, in terms of time after injury, to

avoid zero-time bias when evaluating outcomes

Inclusion/exclusion criteria encompassing time after

injury

Confounders Documentation of and/or study design accounting for:- Any effect of alcohol and/or drugs at time of injury

- Pain

- Comorbidity (eg, preexisting psychiatric morbidity)

- Other possible confounders relevant to the study in

question

Consideration of likely confounders via 1 or more of:

- Amnestic information, direct observation, breath testing,

blood alcohol levels, documentation of concurrent

medication, previous medical history

Use of controls from same population as cases

- For example, patients with orthopedic injuries of the

extremities

- Document how it is established that controls do not also

have MTBI.

Sample size Realistic power calculations at the planning stage Study start only if realistic chance of recruiting sufficient

patients. Multicenter studies may be needed.

Selection

bias and

attrition

Careful balance between the need for clearly defined study

populations and control for confounding, against

negative effects of very narrow inclusion criteria in

terms of limited generalizability

Documentation of numbers and characteristics of

nonparticipants

Consider staffing for the study during peak periods of MTBI

occurrence (weekends, nights).

Outcome

measures

Choice of measures with satisfactory psychometric

properties

Reporting of validity, reliability, and responsiveness of

measures used

Choice of measures assessing clinically meaningful

outcomes specific to MTBI

Transparent consideration of what defines a “clinically

meaningful outcome” within the framework of the study

in question

Relate chosen outcome measures to components of the

International Classification of Functioning, Disability and

Health.

Follow-up Sufficiently long and sufficiently frequent follow-up to

allow evaluation of the outcome in question

Studies with long-term follow-up of over 12mo will be

necessary for assessment of some outcomes.

NOTE. This list is designed to help researchers avoid common pitfalls observed in studies of MTBI 2001e2012. The reader is referred to CONSORT and

STROBE guidelines for more exhaustive, general comments on study design and reporting. The focus for this checklist is on aspects of study design

specific to MTBI.


Externally validated prediction rulesClinical prediction rules are mathematical tools intended to guideclinicians in their everyday decision making.27 Our review found 1prediction rule for long-term recovery after MTBI, and it has notbeen externally validated.28 This is disappointing because it re-flects the likelihood that existing prognostic information is notbeing used to inform patient care. Hence, the development ofclinical prediction rules for recovery post-MTBI is also considereda high priority for future research.

Dynamic prognosis and Bayesian analysisUsual prognosis studies estimate associations between potentialprognostic factors measured at first contact with the health care


system, and the later occurrence of outcomes. This approach isclinically relevant but fails to capture the complexity of the naturalhistory of MTBI and the evolution of its potential prognostic fac-tors.29 Ideally, prognosis should be reconsidered whenever there iscontact with health care providers, especially when symptoms arepersisting. A dynamic prognosis assessment then becomes part ofmonitoring disease progression.30,31 In that monitoring perspective,cohorts are the most appropriate study design, as they includefollow-up. The analysis of such studies can be based on propor-tional hazard models with time-dependent variables, but results ofsuch modeling are often difficult to interpret causally. Promisingapproaches are based on the use of Bayes’ theorem, where theprobability of occurrence of an outcome of interest is estimated



every time new information is available on relevant predictors.30,31

Further research is needed to define the most useful information toinclude in Bayesian models, and to transform Bayesian results intodynamic decision rules accessible for use in clinical practice.

Research Priorities

According to the findings of the ICoMP, the field of MTBI prog-nosis research has advanced slowly. In 2004, the WHO Task Forcelisted 5 research priority areas: (1) the use of confirmatory designs;(2) the influence of pain, psychological distress, and alcohol ordrug abuse on recovery; (3) studies in the elderly; (4) exploration ofthe role of compensation/litigation issues; and (5) further exami-nation of the long-term consequences of MTBI in very youngchildren.1 Most of these are still relevant priorities a decade later.

We prioritized the evidence on prognostic factors using aframework described by Cote et al.32 Phase I studies are hypothesis-generating investigations that explore associations between poten-tial prognostic factors and disease outcomes in a descriptive orunivariate way. Phase II studies are exploratory analyses that focuson particular sets of prognostic factors to discover which factorshave high independent associations with outcomes. Phase IIIstudies are confirmatory with explicit prestated hypotheses thatallow for a focused examination of the strength, direction, and in-dependence of the relationship between a prognostic factor and theoutcome of interest. Prognostic information from accepted phase IIIstudies is considered the strongest evidence of prognostic value,followed by evidence from accepted phase II studies. Phase Istudies do not consider independence of associations or con-founding factors, and they are considered the most preliminary,hypothesis-generating studies. Twenty percent of the acceptedstudies were either randomized controlled trials or phase IIIconfirmatory studies. This is a 17% increase from the 2004 review.1

However, more phase III confirmatory evidence is needed to ensureclinical recommendations are free from confounding influences.

The 2004 WHO Task Force review could not rule out the pos-sibility that observed deficits on cognitive testing might be related toinjury-related pain or psychological distress.1 That issue has nowreceived some attention, and although the number of studies on thistopic is limited, the current evidence suggests that cognitive findingsare unrelated to pain intensity, anxiety, and posttraumatic stressdisorder.33 Not a single accepted article focused on prognosis afterMTBI in the elderly. Given the aging population of most nations, thisis an urgent research priority. There is also no additional informationavailable on litigation/compensation factors in MTBI recovery.Given the consistent evidence that compensation/litigation is asso-ciated with recovery after MTBI, this needs further examination tounderstand motivations behind this association and to suggest pol-icies that promote recovery. Research evidence suggests that childrenwith intracranial pathology on neuroimaging may have a higher riskof postconcussion symptoms and cognitive deficits 1 year post-injury.34 However, well-designed, long-term confirmatory studies areneeded to determine whether this persists for more than 1 year.

In addition, the ICoMP highlighted 10 further research priorityareas. First, measurement studies to determine the diagnostic accuracy,validity, reliability, and responsiveness of definitions for both MTBIand important outcomes are vital. Second, we recommend inclusion oforthopedic injury comparison groups to determine whether subjectivesymptomsare specific to theMTBI.Third, validated clinical predictionrules are needed to identify thosewith prolonged recovery and to targetthe development of interventions to those that might benefit. Fourth, in

order to generalize results to the entire MTBI population, studies needto use participants beyond just hospital admissions or ED patients.These populations capture only the most severe MTBI cases. Fifth,there is little information on the effects of repeated concussion, espe-cially associated with neurodegenerative disorders. Additionally,many of the outcomes use neuropsychological testing, yet under-standing of the clinical importance of changes in neuropsychologicaltesting is limited.15 Sixth, the ICoMP found no acceptable studies onreturn to school. Seventh, to assess the long-termprognosis ofMTBI inthe military, studies need to follow-up injured subjects over longerperiods and assess the interaction of MTBI and posttraumatic stressdisorder. Eighth, we need more studies addressing the differencesbetween adults and children. Ninth, in studies of Parkinson’s disease,researchers need to account for the likelihood that Parkinson’s diseasemay lead to head injury because of changes in cognition or motorcontrol. Tenth, recall bias is problematic when MTBI is based onsubjects’ recall of past concussions.35 Former National Football Lea-gue players who reported more concussions sustained during theirplaying careers in a 2010 survey than in a 2001 survey also complainedof more cognitive deficits on follow-up than those who recalled thesame number of concussions in 2010 and 2001.36 Finally, investigatorsconducting MTBI intervention research should carefully select in-clusion and exclusion criteria, focus on modifiable prognostic factors,and choose the most appropriate outcome measure.

Conclusions

Future research on MTBI prognosis needs to address the issueshighlighted in this review. Standard criteria for defining MTBIwould improve the comparability of studies, but since this currentlydoes not exist, authors need to clearly describe their MTBI criteria.Potential biases, including issues of confounding, information, andselection bias, should be considered in the design phase of aresearch study. Research designs should be selected as appropriatefor the research question, and authors must clearly report theirfindings according to the STROBE and CONSORT guidelines.Studies must include a sample size large enough to address theresearch question with enough statistical power to detect mean-ingful clinical effects. Finally, this article has highlighted a numberof research priorities for further understanding the prognosis ofMTBI including, but not limited to, the following: (1) the use ofconfirmatory designs; (2) studies of measurement validity; (3) focuson the elderly; (4) attention to litigation/compensation issues; (5)the development of validated clinical prediction rules; (6) the use ofMTBI populations other than hospital admissions; (7) continuedresearch on the effects of repeated concussions; (8) longer follow-up times with more measurement periods in longitudinal studies;(9) an assessment of the differences between adults and children;and (10) an account for reverse causality and differential recall.

Keywords

Brain concussion; Epidemiology; Injuries; Methods; Rehabilitation;Research

Corresponding author

Vicki L. Kristman, PhD, Department of Health Sciences, Lake-head University, 955 Oliver Rd, Thunder Bay, Ontario, CanadaP7B 5E1. E-mail address: [email protected].


mailto:[email protected]



Acknowledgments

We thank the other members of ICoMP: Jean-Luc af Geijerstam,MD, PhD; Eleanor Boyle, PhD; Victor G. Coronado, MD, MPH;Pierre Cote, DC, PhD; Cesar A. Hincapie, DC, MHSc; Ryan Hung,MD, MSc; Michelle Keightley, PhD; Alvin Li, BHSc; ConnieMarras, MD, PhD; Peter Rumney, MD; and Britt-Marie Stalnacke,MD, PhD; Panos Lambiris, MSc, Information Scientist, UniversityHealth Network, for assisting in developing, testing, and updatingthe search strategies; and Meijia Zhou, BSc, for assistance withretrieving and screening articles.

References

1. Carroll LJ, Cassidy JD, Holm L, Kraus J, Coronado VG, WHO

Collaborating Centre Task Force on Mild Traumatic Brain Injury.

Methodological issues and research recommendations for mild trau-

matic brain injury: the WHO Collaborating Centre Task Force on Mild

Traumatic Brain Injury. J Rehabil Med 2004;36(43 Suppl):113-25.

2. Cancelliere C, Cassidy JD, Li A, Donovan J, Cote P, Hincapie CA. Sys-

tematic search and review procedures: results of the International Collab-

oration onMildTraumaticBrain Injury Prognosis. ArchPhysMedRehabil

2014;95(3 Suppl 2):S101-31.

3. TheAmericanCongress ofRehabilitationMedicine.MildTraumaticBrain

Injury Committee of the Head Injury Interdisciplinary Special Interest

Group of theAmericanCongress of RehabilitationMedicine. Definition of

mild traumatic brain injury. J Head Trauma Rehabil 1993;8:86-7.

4. National Center for Injury Prevention and Control. Report to Congress

on mild traumatic brain injury in the United States: steps to prevent a

serious public health problem. Atlanta: Centers for Disease Control

and Prevention; 2003.

5. Ruff RM, Iverson GL, Barth JT, Bush SS, Broshek DK, NAN Policy

and Planning Committee. Recommendations for diagnosing a mild

traumatic brain injury: a National Academy of Neuropsychology ed-

ucation paper. Arch Clin Neuropsychol 2009;24:3-10.

6. Williams DH, Levin HS, Eisenberg HM. Mild head injury classifica-

tion. Neurosurgery 1990;27:422-8.

7. Jacobs B, van Ekert J, Vernooy LPL, et al. Development and external

validation of a new PTA assessment scale. BMC Neurol 2012;12:69.

8. Stuss DT, Binns MA, Carruth FG, et al. The acute period of recovery

from traumatic brain injury: posttraumatic amnesia or posttraumatic

confusional state? J Neurosurg 1999;90:635-43.

9. Menon DK, Schwab K, Wright DW, Maas AI, Demographics and

Clinical Assessment Working Group of the International and Interagency

Initiative toward Common Data Elements for Research on Traumatic

Brain Injury and Psychological Health. Position statement: definition of

traumatic brain injury. Arch Phys Med Rehabil 2010;91:1637-40.

10. McCrory P, Meeuwisse W, Johnston K, et al. Consensus statement on

concussion in sport: the 3rd International Conference on Concussion

in Sport held in Zurich, November 2008. Br J Sports Med 2009;

43(Suppl 1):i76-84.

11. Puljula J, Cygnel H, Makinen E, et al. Mild traumatic brain injury

diagnosis frequently remains unrecorded in subjects with craniofacial

fractures. Injury 2012;43:2100-4.

12. Chen AY, Colantonio A. Defining neurotrauma in administrative data

using the International Classification of Diseases Tenth Revision.

Emerg Themes Epidemiol 2011;8:4.

13. World Health Organization. ICD-11 beta draft. February 2013.

Available at: http://apps.who.int/classifications/icd11/browse/f/en#.

Accessed February 14, 2013.

14. Cassidy JD, Cancelliere C, Carroll LJ, et al. Systematic review of self-

reported prognosis in adults after mild traumatic brain injury: results

of the International Collaboration on Mild Traumatic Brain Injury

Prognosis. Arch Phys Med Rehabil 2014;95(3 Suppl 2):S132-51.

15. Iverson GL, Gaetz M. Practical considerations for interpreting change

following brain injury. In: Lovell MR, Echemendia RJ, Barth JT,


Collins MW, editors. Traumatic brain injury in sports. Lisse: Swets &

Zeitlilnger; 2004. p 323-55.

16. Luoto TM, Tenovuo O, Kataja A, Brander AB, Ӧhman J, Iverson GL.

Who gets recruited in mild traumatic brain injury research? J Neu-

rotrauma 2013;30:11-6.

17. McKee AC, Cantu RC, Nowinski CJ, et al. Chronic traumatic en-

cephalopathy in athletes: progressive tauopathy after repetitive head

injury. J Neuropathol Exp Neurol 2009;68:709-35.

18. Boyle E, Cancelliere C, Hartvigsen J, Carroll LJ, Holm LW, Cassidy JD.

Systematic review of prognosis after mild traumatic brain injury in the

military: results of the InternationalCollaboration onMildTraumaticBrain

Injury Prognosis. Arch Phys Med Rehabil 2014;95(3 Suppl 2):S230-7.

19. von Elm E; STROBE Initiative. The Strengthening the Reporting of

Observational Studies in Epidemiology (STROBE) statement: guide-

lines for reporting observational studies. Lancet 2007;370:1453-7.

20. Schulz KF, Altman DG, Moher D, CONSORT Group. CONSORT

2010 statement: updated guidelines for reporting parallel group

randomised trials. BMJ 2010;340:c332.

21. Katz MH. Setting up a multivariable analysis. In: Katz MH, editor.

Multivariable analysis: a practical guide for clinicians. 2nd ed. New

York: Cambridge University Pr; 2006. p 73-95.

22. Shukla D, Devi BI, Agrawal A. Outcome measures for traumatic brain

injury. Clin Neurol Neurosurg 2011;113:435-41.

23. Wilde EA, Whiteneck GG, Bogner J, et al. Recommendations for the

use of common outcome measures in traumatic brain injury research.

Arch Phys Med Rehabil 2010;91:1650-60.

24. Lee YB, Kwon SJ. A more detailed classification of mild head injury in

adults and treatment guidelines. JKoreanNeurosurgSoc 2009;46:451-8.

25. Stucki G, Cieza A, Melvin J. The International Classification of Func-

tioning, Disability and Health (ICF): a unifying model for the conceptual

description of the rehabilitation strategy. J Rehabil Med 2007;39:279-85.

26. Mokkink LB, Terwee CB, Gibbons E, et al. Inter-rater reliability of the

COSMIN (Consensus-based Standards for the selection of health

status Measurement Instruments) checklist. BMC Med Res Methodol

2010;10:82.

27. AdamsST, Leveson SH.Clinical prediction rules. BMJ 2012;344:d8312.

28. Stulemeijer M, van der Werf S, Borm GF, Vos PE. Early prediction of

favourable recovery 6 months after mild traumatic brain injury.

J Neurol Neurosurg Psychiatry 2008;79:936-42.

29. Young NH, Andrews PJD. Developing a prognostic model for trau-

matic brain injuryda missed opportunity? PLoS Med 2008;5:1186-8.

30. Verduijn M, Peek N, Rosseel PM, de Jonge E, de Mol BA. Prognostic

Bayesian networks I: rationale, learning procedure, and clinical use.

J Biomed Inform 2007;40:609-18.

31. Verduijn M, Rosseel PMJ, Peek N, de Jonge E, de Mol BA. Prognostic

Bayesian networks II: an application in the domain of cardiac surgery.

J Biomed Inform 2007;40:619-30.

32. Cote P, Cassidy JD, Carroll L, Frank JW, Bombardier C. A systematic

review of the prognosis of acutewhiplash and a new conceptual framework

to synthesize the literature. Spine (Phila Pa 1976) 2001;26:E445-58.

33. Carroll LJ, Cassidy JD, Cancelliere C, et al. Systematic review of the

prognosis after mild traumatic brain injury in adults: cognitive, psy-

chiatric, and mortality outcomes: results of the International Collab-

oration on Mild Traumatic Brain Injury Prognosis. Arch Phys Med

Rehabil 2014;95(3 Suppl 2):S152-73.

34. Hung R, Carroll LJ, Cancelliere C, et al. Systematic review of the

clinical course, natural history, and prognosis for pediatric mild

traumatic brain injury: results of the International Collaboration on

Mild Traumatic Brain Injury Prognosis. Arch Phys Med Rehabil 2014;

95(3 Suppl 2):S174-91.

35. Godbolt AK, Cancelliere C, Hincapie CA. Systematic review of the

risk of dementia and chronic cognitive impairment after mild trau-

matic brain injury: results of the International Collaboration on Mild

Traumatic Brain Injury Prognosis. Arch Phys Med Rehabil 2014;95(3

Suppl 2):S245-56.

36. Kerr ZY, Marshall SW, Guskiewicz KM. Reliability of concussion

history in former professional football players. Med Sci Sports Exerc

2012;44:377-82.

http://refhub.elsevier.com/S0003-9993(13)01070-8/sref1





http://refhub.elsevier.com/S0003-9993(13)01070-8/sref10a






































http://apps.who.int/classifications/icd11/browse/f/en#













































































Methodological Issues and Research Recommendations for Prognosis After Mild Traumatic Brain Injury:...

Documents

Transcript of Methodological Issues and Research Recommendations for Prognosis After Mild Traumatic Brain Injury:...