Central or peripheral by Kirk Andrew

Archives of Clinical Neuropsychology 23 (2008) 479–485

Commentary

Central or peripheral? A positional stance in reaction to the Prague statement on the role of neuropsychological assessment in sports concussion management Ann B. Shuttleworth-Edwards ∗ National Sports Concussion Initiative (NSCI), Psychology Department, Rhodes University, Grahamstown, South Africa Accepted 20 July 2008

Abstract The objective of this paper is to clarify the role of neuropsychological assessment in sports concussion management in response to the Prague summary statement of the most recent international conference on concussion (McCrory, P., Johnston, K., Meeuwisse, W., Aubry, M., Cantu, R., Dvorak, J., et al. (2005). In the Prague statement it is proposed that there should be a new categorization of concussion (‘simple’ versus ‘complex’) to inform management of the injury, including the suggestion that neuropsychological assessment is not indicated for ‘simple’ concussion. However, from a neuropsychological perspective the Prague position is conceptually problematic. The present paper proposes on theoretical and empirical grounds, that neurocognitive evaluation is warranted for any concussive injury to increase diagnostic sensitivity and provide prognostic indications. Further, three levels of neurocognitive assessment and their professionally legitimate implementation (that are poorly defined in the Prague paper) are delineated. Recommendations are provided for consideration at the next international concussion meeting, with emphasis on giving due weight to the neuropsychological position. © 2008 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved. Keywords: Sports concussion management; Computerized tests; Professional issues

1. Introduction The objective of this paper is to provide a positional commentary on a number of concerns of relevance to the discipline of clinical neuropsychology that arise out of the summary statement of the Prague second international conference on concussion (McCrory et al., 2005). Specifically the issues to be addressed are (i) a proposed twoway categorization of concussion (‘simple’ versus ‘complex’) and the associated suggestion that neuropsychological assessment is not indicated for ‘simple’ concussion, and (ii) types of neurocognitive assessment and their professionally appropriate implementation (that are poorly defined in the Prague paper). To the author’s knowledge, despite several years having elapsed, there has been no formal evaluation of the practice directives contained in the Prague statement from a neuropsychological perspective. A positional response of this type appears urgent given the imminent third ∗ Correspondence address: NSCI, Psychology Clinic, Rhodes University, Grahamstown 6139, South Africa. Tel.: +27 46 6361296; fax: +27 46 6361296. E-mail address: a.edwards@ru.ac.za.

480

A.B. Shuttleworth-Edwards / Archives of Clinical Neuropsychology 23 (2008) 479–485

international conference on concussion during which continued debate and recommendations on these matters will be resumed. 2. Neuropsychological testing in concussion management In recent years two conferences have been convened, bringing together a range of experts to provide recommendations for improved safety and health of athletes who sustain concussive injuries in sports (Aubry et al., 2002; McCrory et al., 2005). Each provided a summary and agreement statement designed to establish principles on the issue of sports concussion and its optimal management known as the Vienna and Prague summary papers, respectively. Within the Vienna statement (Aubry et al., 2002) two revolutionary guiding principles for concussion management were established that constituted a significant advance when compared with prior rule-based practice dependant on concussion grading systems. These were as follows: (i) No single concussion grading system was endorsed as a scientifically valid basis on which to inform return-to-play decisions, and therefore an individualized clinical approach to injury recovery and decisions about return-to-play was recommended; (ii) Neuropsychological testing including paper-andpencil and/or computerized test programmes was declared to be one of the cornerstones of this clinical assessment, contributing substantially to understanding the injury and to the overall medical management of the individual. No differentiation was made as to which athletes would require such evaluation, and given the emphasis on the individualized approach, the logical assumption is that this would apply to every concussed athlete. Three years later in the Prague summary statement (McCrory et al., 2005) both of the aforementioned guiding principles from the original conference were broadly endorsed, as follows: (i) “The Vienna recommendation that injury grading scales be abandoned in favour of combined measures of recovery to determine injury severity (and/or prognosis) and hence individually guide return-to-play decisions received continued support” (p. 49); and (ii) “The application of neuropsychological testing in concussion has been shown to be of value and continues to contribute significant information in concussion evaluation” (p. 51). However, in contrast to the Vienna statement, the Prague statement does propose a differentiation as to how and when concussed athletes should be evaluated, based on a newly proposed two-way categorisation of ‘simple’ versus ‘complex’ concussion. This embodies a surprising shift in emphasis that substantially diminishes the centrality of neuropsychological testing in concussion management, and therefore warrants careful evaluation. 3. Simple versus complex concussion: an anomalous proposal for management? As indicated earlier, in the Prague paper (McCrory et al., 2005) it is stated that the old grading scales have been ‘abandoned in favor of combined measures of recovery in order to determine injury severity (and/or prognosis) and hence individually guide return-to-play decisions’ (p.49). However, in light of the above-stated ‘abandoned’ approach, an apparent conceptual anomaly appears in the paper with the introduction of a new classification system of ‘simple’ versus ‘complex’ concussion to inform post concussion management (p.50). It is not clear how this attempt at classification differs fundamentally from the prior grading scales, being in essence yet another grading system without scientific validation upon which to base concussion management indications. Specifically, simple concussion is defined as an injury that progressively resolves without complication over a period of 7–10 days, and it is stated that ‘mental status screening’ rather than ‘formal neuropsychological screening’ is necessary in monitoring recovery in these cases that account for most instances of sports concussive events. Complex concussion is defined as an injury which results in persistent symptoms, specific neurological sequelae such as convulsions or loss of consciousness (>1 min), or prolonged cognitive impairment. This group may also include individuals who have sustained repeated concussions over time, or suffer concussions with increasingly less impact force, and for this group neuropsychological assessment is deemed appropriate. In the wake of the Prague statement, there appears to be only one study to date that has attempted to verify the proposed simple–complex delineation (Iverson, 2007b). In the research report Iverson challenges the notion reflected in the Prague statement that most concussions are of the ‘simple’ type, suggesting that this is an observation that appears to be loosely based on the recovery of the older athlete only, and does not take into account the recovery patterns of the younger athlete. Further, Iverson demonstrated that concussion history did not influence time to recovery, and therefore failed to validate the use of this variable for the simple–complex classification. Correspondence arose in relation to the Iverson challenge (Iverson, 2007a; Johnston & McCrory, 2007; Meeuwisse, 2007), in the process of which it was acknowledged that refinement on the simple–complex issue is called

A.B. Shuttleworth-Edwards / Archives of Clinical Neuropsychology 23 (2008) 479–485

481

for. Whilst this is an encouraging concession, ‘refinement’ carries with it the implication that the fundamental concept may be acceptable. However, notwithstanding the lack of empirical validation of the simple–complex division, the notion contains an inherent conceptual flaw that relates to differential diagnosis, as follows. According to a relatively coarse cognitive sideline measure such as the Sideline Concussion Assessment Tool (SCAT) (McCrory et al., 2005), an individual may reveal apparent resolution of symptoms within the 7–10-day cut-off period thereby warranting the diagnosis of a ‘simple’ concussion. However, this same individual may well demonstrate the presence of significant persistent neurocognitive dysfunction via a more challenging computer-based assessment, albeit this form of assessment would specifically not be made available to them (according to the Prague recommendations) because their injury falls within the category of a ‘simple’ concussion. In particular, a relatively protected individual in terms of cognitive reserve (Satz, 1993; Stern, 2002) (i.e. a high level of intellectual ability and the absence of neurological or psychiatric vulnerability) will require more stringent cognitive testing in order to reveal cognitive signs following a concussion, that might easily be missed on the basis of the cursory mental status screening. Hence, in order to diagnose a ‘simple’ concussion it is necessary to confirm the absence of symptoms by virtue of the most effective neurocognitive assessment, and it is therefore anomalous to suggest that the diagnosis in itself can dictate whether or not such assessment is needed in the first place. Iverson (Iverson, 2007b, p. 36 ) endorses the difficulty as follows: “However, what if the asymptomatic athlete has apparent balance problems based on the Balance Error Scoring System (BESS) or dynamic posturography? We would not know unless we tested them, so we face somewhat of a conundrum. Do we rely entirely on athletes’ self-report, or do we include other outcome measures, such as balance testing and neuropsychological testing, in the clinical management protocol?” In the final analysis the term ‘simple’ is potentially misleading and has overtones of the disfavoured expression in clinical neuropsychology of ‘minor’ head injury, that may result in overlooking significant difficulties that are known to occur for a percentage of individuals following even the single mild TBI (Boll, 1983; Ruff, 2005; Reitan & Wolfson, 1999). Fine neuropsychological screening would appear particularly necessary in such instances of the ostensibly mildest of traumatic brain injury, in order to ensure that any residual cognitive sequelae have been identified. Empirically this point has been elegantly demonstrated in a study of high school athletes diagnosed with Grade 1 concussion (i.e. symptom resolution within 15 min), who would normally have been returned to play in the game (Lovell, Collins, Iverson, Johnston, & Bradley, 2004). For the purposes of this study, a group of such athletes were kept out of further play, and when tested on a sensitive computerized test battery, revealed neurocognitive deficits up to 3 days post injury. A recent critical study similarly demonstrates that a group of high school and collegiate athletes who denied the presence of symptoms within one week of a concussion were impaired on all four neurocognitive composite scores of the ImPACT battery (Fazio, Lovell, Pardini, & Collins, 2007). Finally, a number of recent case controlled studies have endorsed the increased diagnostic accuracy of sports-related concussion via the inclusion of computerized neurocognitive testing as part of the overall clinical assessment of the athlete (Broglio, Macciocchi, & Ferrarra, 2007; Van Kampen, Lovell, Pardini, Collins, & Fu, 2006). From these studies it is clear that reliance on the symptomatic reports of the athlete themselves increases the potential for false negative diagnosis, and the associated risk of premature return-to-play. Thus in sum, on both conceptual and empirical grounds, there is compelling evidence to suggest that the Prague statement does not give adequate credence to the need for fine neurocognitive testing, especially for the elusive so-called ‘simple’ cases, where it has crucial relevance to ensure that ‘silent’ symptoms have not been missed with potentially deleterious clinical consequences. It would appear, therefore, that routine neuropsychological testing following every concussion, over and above coarse mental status screening, will reduce the potential for false negative diagnosis, and be commensurate with best medical practice. It is important to note that best practice in this instance, does not imply mandatory practice, but rather a recommended goal towards which it is possible for sports practitioners to aspire in light of current research indications. 4. Modes of neuropsychological assessment and their application Another source of confusion in the Prague document is lack of definition of what is meant by a number of terms used in the document including ‘mental status screening’, ‘cognitive assessment’, ‘neuropsychological screening’, ‘formal neuropsychological assessment’. To gain clarity in this situation, it is useful to conceptualize ‘neuropsychological assessment’ as an umbrella term, being defined as a wide-ranging investigation that involves the assessment and

482

A.B. Shuttleworth-Edwards / Archives of Clinical Neuropsychology 23 (2008) 479–485

integration of cognitive as well as emotional and behavioural sequelae of brain impairment. Similarly, ‘mental status screening’ can be conceptualized as a process that involves a broad-based (albeit brief) evaluation of an individual’s cognitive, emotional, and behavioural status. These terms are contrasted with the more speciﬁc process of ‘neurocognitive assessment’ that may occur as part of the broader neuropsychological assessment or mental status examination at various levels of psychometric test sophistication. Accordingly, in operational terms, three levels of neurocognitive evaluation apply including: Level 1: Checklist neurocognitive screening via a relatively gross sideline type assessment especially devised to identify the presence of any immediate acute sequelae of a concussive episode, involving relatively all-or-none responses at a low level of psychometric sophistication (e.g. the Sideline Concussion Assessment Tool—SCAT) (McCrory et al., 2005); Level 2: Computerized neurocognitive testing via one of the computer-based programmes especially developed for baseline testing and follow up of subacute sports concussion sequelae (for example ANAM, CogSport, Headminder and ImPACT) that produces computer-generated scores for a variety of cognitive modalities (depending on the particular programme in question) and is usually statistically contrasted with baseline and/ or normative data on an automatically generated report; Level 3: Comprehensive neuropsychological testing that may include selected computerized tests, but also typically incorporates a comprehensive battery of traditional psychometric tests (for example the Wechsler intelligence and memory scales; selected paper-and-pencil tests to probe specific functions such as the Trail Making Test, Rey Complex Figure and the Stroop Test), the qualitative and quantitative results of which are integrated in relation to clinical and normative data to inform a neuropsychodiagnostic opinion. This third form of assessment would be called for in complicated cases of concussion, that do not recover after six to eight weeks, to evaluate the need for further neurological investigation or recommendations about termination of a sporting career that has high risk of concussive injury. Different implications apply to these three types of assessment as defined above, in terms of their appropriate administration and interpretation. At the first and third levels of assessment, the indications are relatively clear-cut. The sideline checklist, which is at the lowest level of test sophistication, can be administered by medical or sports personnel, who require only a rudimentary level of training for its administration and the meaning of the test outcome. On the other hand, the in-depth specialist neuropsychological assessment which is at the top level of sophistication, requires the services of a registered psychologist trained in psychometrics to integrate the meaning of a spectrum of test data in terms of brain–behaviour relationships and neuropsychiatric diagnosis. However, it is the middle level of assessment, involving the use of computerized neurocognitive tests especially designed for the sports concussion arena, where the correct professional stance to administration and interpretation is less clear-cut and the topic of some debate. 5. Computerized neuropsychological assessment: the issue of lawful practice Both the Vienna and Prague summary statements note that computerized testing has the advantage over traditional neuropsychological test batteries in that it may be administered by a team doctor or be web-based rather than needing a neuropsychologist for a formal assessment (Aubry et al., 2002; McCrory et al., 2005), with the implication that there is no need for the neuropsychologist in the assessment process. However, this commentary does not suitably deal with the situation, in that it skirts the issue of interpretation, and in this regard more detailed debate is available in the literature. On the one hand it has been suggested that “ideally, a neuropsychologist would administer and interpret the results of every neuropsychological test conducted in sports medicine” (Collie & Maruff, 2003), but that this is not necessary when automated computerized reports are available in that they provide the facility for simple ‘yes’ or ‘no’ answers as to whether cognitive function has changed or not following concussion. This article goes on to suggest that the practical constraints of exceedingly large numbers of athletes render it unreasonable to expect that administration and interpretation of computerized neuropsychological tests can be carried out exclusively by neuropsychologists, and given the automated aspect of these tests, the proposal is that their independent use by medical practitioners is legitimate. In contrast, others have expressed concern about the potential for misuse of neuropsychological tests with psychometric properties if they become separated from their professional – i.e. neuropsychological – source (Shuttleworth-Edwards & Border, 2002; Shuttleworth-Edwards & Whitefield, 2007a; Shuttleworth-Edwards & Whitefield, 2007b). These authors warn that the ease with which computerized tests can be administered, and automatic reports generated, may cause non-psychologists to fall into the trap of construing that the scores derived can be used simplistically as a type

A.B. Shuttleworth-Edwards / Archives of Clinical Neuropsychology 23 (2008) 479–485

483

of ‘yes’ or ‘no’ litmus paper test for making decisions about the presence or absence of cerebral dysfunction in the individual case. Such a cookbook approach to the interpretation of psychometric test data is reminiscent of the dark ages of actuarial neuropsychology, and has been emphatically discredited in modern approaches to clinical neuropsychology (Lezak, Howieson, & Loring, 2004; Shuttleworth-Jordan, 1997; Walsh, 1991). In the modern approach, test data are not deemed to have absolute qualities, and for a valid interpretation it is expected that they should be: (i) contextualized in relation to the influential variables of a testee’s age, gender, race, language, level and quality of education and intellectual potential; (ii) understood in relation to the nature of the injury, possible physical and mental co-morbidity, test-taking conditions, and test-specific psychometric properties; and finally (iii) evaluated for patterns suggestive of typical test artifacts, functional rather than organic pathology, and malingering, all of which need to be excluded. Multiple case examples (that are unfortunately beyond the scope of the present commentary) can be produced to illustrate how failure to do this may be hazardous, in that any one of these factors alone or in combined form may easily result in false positive or false negative diagnosis and consequent mismanagement. It would be a travesty if the substantial advance afforded by the new technology in the form of computerized neurocognitive testing in the sports arena becomes the source of a return to the dark ages of cookbook neuropsychology. In short, the principle of interpretation of psychometric test scores remains the same whether or not they have been computer generated, and calls for the refined level of understanding available to the registered psychologist with training in psychometrics and neuropsychological diagnostic skills. In accordance with this more restricted professional outlook, the use of a psychologist for the purposes of psychometric assessment is advocated in professional position statements in both the USA and South Africa (Guskiewicz et al., 2004; Makgoke, 2004; Moser et al., 2007). In a statement from the National Athletic Trainers’ Association (USA) it is pointed out that most states require advanced training and licensure to purchase and use neuropsychological tests for clinical purposes, and at present these requirements necessitate that a licensed psychologist oversee and supervise the clinical assessment (Guskiewicz et al., 2004). This position has been broadly endorsed specifically in relation to sports concussion assessment in the USA, in the forum of the National Academy of Neuropsychology (Moser et al., 2007). Similarly, the Health Professions Council for South Africa (HPCSA) states that: “Psychological tests have to be administered, scored, interpreted and reported on by registered persons, who have recognized and appropriate education and training in the field of psychometry” (Makgoke, 2004). Any deviation from this policy is considered illegal and needs to be reported for immediate prosecution (Makgoke, 2004). In South Africa, with specific reference to computerized testing, it is indicated that whilst these may be administered by suitably trained non-psychologists, this must still be under the supervisory control of a registered psychologist1 (Health Professions Council of South Africa (HPCSA): Professional Board of Psychology, 2006). Further, HPCSA policy makes it clear that the interpretation of the computerized test results, feedback and reporting need to be undertaken by a registered psychologist. Despite these expressed policies, it appears that in many settings, including in the USA and South Africa, the computerized neuropsychological evaluation is being conducted by sports medicine clinicians without any consultation with neuropsychologists on the basis of one or more of the reasons already indicated above. Firstly, there is the underlying premise that computerized assessment can be conducted without the involvement of a psychologist or neuropsychologist (i.e. the discredited cookbook approach). Alternatively, lip service is given to the fact that the involvement of a neuropsychologist is optimal, but this is deemed to be an impractical option for treating large numbers of athletes and/or is considered not to be viable in places where there is limited access to specialist psychological services. A concern is that these arguments may be advanced as a means of defending a position which allows medical doctors and other practitioners to perform skills which they are not trained in and are the specialist field of the registered psychologist or neuropsychologist. However, a key feature of computerized cognitive tests is that they facilitate mass testing and at the same time allow for the electronic submission of data to a neuropsychologist for interpretation and resubmission back to a trainer or sports doctor from anywhere in the world in a matter of minutes (Collie, Darby, & Maruff, 2001; Shuttleworth-Edwards & Whitefield, 2007b). Under medico-legal scrutiny, the argument of scarcity of appropriate resources (especially when this is not strictly the case) may not be accepted as a basis for justifying less than optimal parameters for professional practice, and offering the resultant lowered, and potentially hazardous standards of care.

In South Africa there is no specialist registration for clinical neuropsychology and hence the term ‘neuropsychologist’ in the South African professional setting refers to a registered psychologist who has developed a special interest area and expertise in neuropsychology.

484

A.B. Shuttleworth-Edwards / Archives of Clinical Neuropsychology 23 (2008) 479–485

6. Recommendations Based on the arguments presented above the following proposals are suggested for consideration during the next international conference on sports concussion: 1. It should be acknowledged that the proposed ‘simple’ versus ‘complex’ categorization of concussion (and its related management recommendations) is an anomalous proposal. It is no different from returning to a new form of the old concussion grading system to inform management that has been appropriately discredited. Therefore, rather than wasting time trying to refine a fundamentally flawed and empirically invalidated approach it should be completely abandoned. 2. In relation to point 1 above, it is proposed that discussions should focus on clearly re-instating the advisability of an individualized approach to concussion management that includes incorporating neurocognitive assessment for any concussive episode regardless of severity as was the consensus opinion of the earlier Vienna meeting. 3. Given that neurocognitive assessment is such a central component of concussion management, substantial weight needs to given to recommendations from within the discipline of Clinical Neuropsychology about what issues have relevance from a neurodiagnostic perspective, as well as what constitutes legitimate practice for the administration and interpretation of neuropsychological tests at the various levels delineated in this paper. This should be extensively debated amongst a suitably representative multidisciplinary team in order to provide a more conceptually coherent consensus statement than arose out of the Prague convention. Great care should be taken in the future that an influential document of this type, that has potentially far-reaching medical consequences, is at a high level of refinement and includes dissenting viewpoints if necessary prior to its publication 4. Finally, as an addendum, it is noted that cross-cultural issues (that are central to neurocognitive assessment) are not dealt with as an issue in the Prague statement. Moreover author affiliation can be challenged on the basis of inadequate international representativeness, being heavily eurocentric. Three authors are from the USA, one from Canada, two from Switzerland, one from Germany, and one from Australia. It is recommended, therefore, that there is an attempt to make a consensus statement arising out of the third concussion conference more truly internationally representative. The consideration of adequate concussion management strategies for relatively disadvantaged populations gives rise to a specific set of challenges that are not necessarily relevant to the elite. Conﬂict of interest The author is commercially involved in the development of computerized neurocognitive screening as a component of sports concussion management in South Africa and the UK, using the ImPACT programme. References Aubry, M., Cantu, R., Dvorak, J., Graf-Baumann, T., Johnston, K. M., Kelly, J., et al. (2002). Summary and agreement statement of the 1st international Symposium on Consussion in Sport, Vienna, 2001. Clinical Journal of Sport Medicine, 12, 6–11. Boll, T. J. (1983). Minor head injury in children: Out of sight but not out of mind. Journal of Clinical Child Psychology, 12, 74–80. Broglio, S. P., Macciocchi, S. N., & Ferrarra, M. S. (2007). Sensitivity of the concussion assessment battery. Neurosurgery, 60, 1050–1058. Collie, A., Darby, D., & Maruff, P. (2001). Computerised cognitive assessment of athletes with sports related head injury. British Journal of Sports Medicine, 35, 297–302. Collie, A., & Maruff, P. (2003). Computerised neuropsychological testing. British Journal of Sports Medicine, 37, 2–3. Fazio, V. C., Lovell, M. R., Pardini, J. E., & Collins, M. W. (2007). The relation between post concussion symptoms and neurocognitive performance in concussed athletes. NeuroRehabilitation, 22, 207–216. Guskiewicz, K. M., Bruce, D. E., Cantu, R. C., Ferrara, M. S., Kelly, J. P., McCrea, M., et al. (2004). National Athletic Trainers’ Association position statement: Management of sport-related concussion. Journal of Athletic Training, 39, 280–297. Health Professions Council of South Africa (HPCSA): Professional Board of Psychology (2006). List of tests classiﬁed as being psychological tests (Rep. No. Form 207). Iverson, G. (2007a). Letter to editor: Reply. Clinical Journal of Sport Medicine, 17, 330. Iverson, G. (2007b). Predicting slow recovery from sport-related concussion: The new simple–complex distinction. Clinical Journal of Sport Medicine, 17, 31–37. Johnston, K. M., & McCrory, P. (2007). Letter to editor. Clinical Journal of Sport Medicine, 17, 330. Lezak, M. D., Howieson, D. B., & Loring, D. W. (2004). Neuropsychological assessment (4th ed.). Oxford: Oxford University Press.

A.B. Shuttleworth-Edwards / Archives of Clinical Neuropsychology 23 (2008) 479–485

485

Lovell, M. R., Collins, M. W., Iverson, G. L., Johnston, K. M., & Bradley, J. P. (2004). Grade 1 or “ding” concussions in high school athletes. Journal of Neurosurgery, 98, 296–301. Makgoke, P. (2004). Media statement by the Health Professions Council of South Africa (HPCSA). South African Psychiatry Review, 7, 39. McCrory, P., Johnston, K., Meeuwisse, W., Aubry, M., Cantu, R., Dvorak, J., et al. (2005). Summary and agreement statement of the 2nd International Conference on concussion in sport, Prague 2004. Clinical Journal of Sport Medicine, 15, 48–55. Meeuwisse, W. H. (2007). The editor-in-chief responds. Clinical Journal of Sport Medicine, 17, 331. Moser, R. S., Iverson, G. L., Echemendia, R. J., Lovell, M. R., Schatz, P., Webbe, F. M., et al. (2007). NAN position paper. Neuropsychological evaluation in the diagnosis and management of sports concussion. Archives of Clinical Neuropsychology, 22, 909–916. Reitan, R. M., & Wolfson, D. (1999). The two faces of mild head injury. Archives of Clinical Neuropsychology, 14, 191–202. Ruff, R. (2005). Two decades of advances in understanding of mild traumatic brain injury. Journal of Head Trauma Rehabilitation, 20, 5–18. Satz, P. (1993). Brain reserve capacity on symptom onset after brain injury: A formulation and review of evidence for threshold theory. Neuropsychology, 7, 273–295. Shuttleworth-Edwards, A. B., & Border, M. A. (2002). Computer based screening in concussion management: use versus abuse. British Journal of Sports Medicine, 36, 473. Shuttleworth-Edwards, A. B., & Whitefield, V. J. (2007a). Ethically we can no longer sit on the fence: A neuropsychological perspective on the cerebrally hazardous contact sports. South African Journal of Sports Medicine, 19, 32–38. Shuttleworth-Edwards, A. B., & Whitefield, V. (2007b). Optimal application of neurocognitive assessment in concussion management: A professional dilemma. South African Journal of Sports Medicine, 19, 101–104. Shuttleworth-Jordan, A. B. (1997). Age and education effects on brain-damaged subjects: Negative findings revisited. The Clinical Neuropsychologist, 11, 205–209. Stern, Y. (2002). What is cognitive reserve? Theory and research application of the reserve concept. Journal of the International Neuropsychological Society, 8, 448–460. Van Kampen, D. A., Lovell, M. R., Pardini, J. E., Collins, M. W., & Fu, F. H. (2006). The “value added” of neurocognitive testing after sports-related concussion. The American Journal of Sports Medicine, 34, 1630–1635. Walsh, K. (1991). Understanding brain damage: A primer of neuropsychological evaluation (2nd ed.). Melbourne: Churchill Livingstone.