Performance across Domains of Attention after Pediatric Traumatic Brain Injury

Material Information

Performance across Domains of Attention after Pediatric Traumatic Brain Injury
Boegehold, Lindsey Michelle
Place of Publication:
[Gainesville, Fla.]
University of Florida
Publication Date:
Physical Description:
1 online resource (46 p.)

Thesis/Dissertation Information

Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Clinical and Health Psychology
Committee Chair:
Heaton, Shelley C.
Committee Members:
Fennell, Eileen B.
Robinson, Michael E.
Eyberg, Sheila M.
Graduation Date:


Subjects / Keywords:
Attention deficit hyperactivity disorder ( jstor )
Child psychology ( jstor )
Craniocerebral trauma ( jstor )
Neuropsychology ( jstor )
Normativity ( jstor )
Parents ( jstor )
Pediatrics ( jstor )
Physical trauma ( jstor )
Selective attention ( jstor )
Traumatic brain injury ( jstor )
Clinical and Health Psychology -- Dissertations, Academic -- UF
attention, brain, cognitive, injury, pediatric
Electronic Thesis or Dissertation
bibliography ( marcgt )
theses ( marcgt )
Psychology thesis, M.S.


Attentional impairments are common after pediatric traumatic brain injury (TBI) and a clear understanding of the nature of these impairments is critical for development of targeted rehabilitation strategies. However, past literature has produced contradicting and inconclusive findings regarding the pattern of impairment across attentional domains. The current study sought to unify past studies using a multidimensional model of attention to characterize the pattern of attentional skills after moderate to severe pediatric TBI. Children ages 6-16 were administered the Test of Everyday Attention for Children (TEA-Ch) within one year of sustaining a moderate/severe TBI (n=26) or orthopedic injury not involving the head (OI; n=18). Subtests of the TEA-Ch were selected to represent performance in four attentional domains: Sustained Attention (Score!), Divided Attention (Score DT), Selective Attention (Sky Search), and Attentional control (Creature Counting). Statistically, Bonferroni-corrected t-tests revealed that the TBI group scored significantly lower than the OI group on tests of Sustained and Divided Attention (Score! total correct, p=.009; Score! DT total correct, p=.003), and a test of Attentional Control (Creature Counting timing score, p=.011). Qualitatively, the TBI group exhibited clinically significant impairment relative to the test normative sample on the task measuring Attentional Control (Creature Counting, Scaled Score=5.00). This study revealed a pattern of either statistically or clinically significant impairment in sustained and divided attention, as well as attentional control after pediatric TBI. Furthermore, average accuracy in the context of impaired speeded performance was observed on timed tasks, suggesting that attentional deficits may be partially explained by impaired motor or cognitive processing speed. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis (M.S.)--University of Florida, 2008.
Adviser: Heaton, Shelley C.
Electronic Access:
Statement of Responsibility:
by Lindsey Michelle Boegehold

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Lindsey Michelle Boegehold. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
LD1780 2008 ( lcc )


This item has the following downloads:

Full Text




2008 Lindsey Michelle Boegehold 2


To B. Abbins 3


ACKNOWLEDGMENTS First and foremost, I would like to thank Dr. Shelley Heaton for her guidance and support from conceptualization of this st udy to finalization of the manuscri pt. I would also like to thank my committee members (Dr. Eileen Fennell, Dr. Michael Robinson, and Dr. Sheila Eyberg) for their helpful advice. This study would not have been possible without the hard work of many graduate and undergraduate research assistants including Cara Kimberg, William Watson, Sarah McCann, and Jenna Alberto. Finally, I would like to thank my friends and family for their love, support, and encouragement. 4


TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................6 LIST OF FIGURES .........................................................................................................................7 ABSTRACT .....................................................................................................................................8 CHAPTER 1 INTRODUCTION................................................................................................................. .10 2 METHODS...................................................................................................................... .......19 Participants .............................................................................................................................19 Procedures ...............................................................................................................................20 Measures .................................................................................................................................20 Parent Report Measures ...................................................................................................20 Child Assessment ............................................................................................................22 Statistical Analysis ..................................................................................................................24 3 RESULTS...................................................................................................................... .........26 Pre-Injury Functioning ............................................................................................................26 Attention Measures .................................................................................................................26 Sustained Attention .........................................................................................................26 Divided Attention ............................................................................................................27 Selective Attention ..........................................................................................................27 Attentional Control ..........................................................................................................28 4 DISCUSSION................................................................................................................... ......32 LIST OF REFERENCES ...............................................................................................................42 BIOGRAPHICAL SKETCH .........................................................................................................46 5


LIST OF TABLES Table page 2-1. Participant demographics .......................................................................................................25 3-1. Parent ratings of pre-injury attentional perfor mance (T-scores) ............................................30 3-2. Performance of children with moderate-s evere TBI or orthopedi c injuries on the TEACh (scaled scores) ..............................................................................................................30 6


LIST OF FIGURES Figure page 3-1. Attentional performance of children with moderate-severe TBI or orthopedic injuries .......31 7


Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science PERFORMANCE ACROSS DOMAINS OF A TTENTION AFTER PEDIATRIC TRAUMATIC BRAIN INJURY By Lindsey Michelle Boegehold May 2008 Chair: Shelley C. Heaton Major: Psychology Attentional impairments are common after pe diatric traumatic brain injury (TBI) and a clear understanding of the nature of these impairments is critical for development of targeted rehabilitation strategies. However, past literature has produced contradicting and inconclusive findings regarding the pattern of impairment across attentional domains. The current study sought to unify past studies using a multidimen sional model of attention to characterize the pattern of attentional skills after m oderate to severe pediatric TBI. Children ages 6 to 16 years were administered the Test of Everyday Attention for Children (TEA-Ch) within one year of su staining a moderate/severe TBI ( n=26) or orthopedic injury not involving the head (OI; n=18). Subtests of the TEA-Ch were selected to represent performance in four attentional domains: Sustained Attent ion (Score!), Divided Attention (Score DT), Selective Attention (Sky Search), and A ttentional control (Creature Counting). Statistically, Bonferroni-correct ed t-tests revealed that the TBI group scored significantly lower than the OI group on tests of Sustained and Divided Attention (Score! total correct, p=.009; Score! DT total correct, p=.003), and a test of Attenti onal Control (Creature Counting timing score, p=.011). Qualitatively, the TBI group exhibi ted clinically significant impairment 8


relative to the test normative sample on the ta sk measuring Attenti onal Control (Creature Counting, Scaled Score=5.00). Results showed a pattern of either statistica lly or clinically significant impairment in sustained and divided attention, as well as atten tional control after pediat ric TBI. Furthermore, average accuracy in the context of impaired speeded performance was observed on timed tasks, suggesting that attentional deficits may be pa rtially explained by impaired motor or cognitive processing speed. 9


CHAPTER 1 INTRODUCTION Pediatric traumatic brain injury (TBI) is a major public health problem in the United States. According to a recent report publishe d by the Centers for Disease Control (CDC; Langlois, Rutland-Brown, & Thomas, 2006), nearly half a million childre n sustain a TBI every year, and about 2,700 will die. Of the survivors, the majority have minor injuries and generally good outcomes. However, about 15-18% of cases involve more severe br ain injuries (Kraus, 1995; Lescohier & DiScala, 1993). Todays medi cal advances have enabled an increasing number of these severely injured children to survive (Sosin, Snieze k, & Waxweiler, 1995). Rehabilitation techniques, however, have lagged behind. Thus, many children who survive a moderate or severe TBI go on to experience perm anent disabilities, such as impaired cognition, behavioral problems, and psychosocial deficits (Keenan & Bratton, 2006; Ewing-Cobbs et al., 2004; Van Heugten et al., 2006). Impaired attention is a comm on complaint following pediatri c TBI (Hooper et al., 2004). This may be in part due to the fact that frontal lobe white matter and prefrontal cortical regions, which are thought to be central components of th e brains attentional system, are some of the most commonly injured areas in pediatric TB I (Mirsky, Anthony, Duncan, Ahearn, & Kellam, 1991; Levin et al., 1997; Wilde et al., 2005). The vulnerability of attention in pediatric brain injury is particularly worrisome, considering the importance of intact attentional skills for many aspects of child functioning and development. Although the long-term consequences of attent ional deficits in pediatric TBI have not been fully investigated, studies of the behavi oral, emotional, and cognitive correlates of attentional deficits in childhood Attention Deficit/Hyperactivity Disorder (ADHD) provide some clues to the potential impact in TBI. Studies have shown th at children diagnosed with ADHD 10


are at higher risk for mood and anxiety disorders, substance dependence, a nd disruptive behavior disorders (Biederman et al., 2006; Wilens, 2004). Furthermore, ADHD has been linked to poor adaptive functioning (Stavro, Ettenhofer, & Nigg, 2007) and low academic achievement (Wilson & Marcotte, 1996). These studies suggest that at tentional problems in childhood, such as those incurred following a pediatric TB I, may put children at risk for the development of psychosocial, behavioral, academic, and adaptive skills difficulties. Despite the importance of understanding and reme diating attentional deficits after TBI, our knowledge of the exact nature of these impairme nts in childhood TBI is somewhat limited. The majority of studies examining attentional functioning followi ng pediatric TBI utilize some variation of a continuous performance task, whic h is thought to measure sustained attention and inhibition of automatic responses. Generally, th is research has shown that children who are injured at a younger age and children who are more severely injured have more deficits in sustained attention (Dennis, Wilkinson, Koski, & Humphreys, 1995; Catroppa & Anderson, 2003; Van Heugten et al., 2006). Attention has been shown to be impaired in the acute stages of pediatric TBI, and for some children, particularly those with more severe injuries, these deficits persist into more chronic stages of recovery (Catroppa, Anderson, Morse, Haritou, & Rosenfeld, 2006). Although generalizations abou t attentional functioning afte r pediatric TBI are important, a more complete understanding of the pattern of attentional impairment can be achieved by working within a multidimensional framework. Attention is typically re ferred to as if it is a unitary construct; however, current neuropsychological and cognitive models suggest that it is comprised of multiple domains (Mirsky et al., 1991; Cooley & Morris, 1990; Po sner & Peterson, 1990). There is significant variation between the numerous proposed models of attention, but most include dimensions such 11


as selective or focused attention, sustained attention, divide d attention, and a ttentional control (i.e., aspects of executive functioning). Understanding attention as a multidimensional construct facilitates a more thorough char acterization of the pattern of attentional impairment following brain injury. This comprehensive characterization is a crucial first step toward designing effective rehabilitation strategies. However, the majority of studies on at tention in pediatric TBI are not based on these current theories of attent ion. Rather, most studies present findings for only one attentional domain, making it difficult to draw conclusions about the pattern of attentional impairment following head injury. Methodological differences across studies have further impeded understanding of attentional skills after pediatric TBI. First, th ere are several developmental variables that can affect a childs performance on measures of cognitive ability after head injury. A brain injury in childhood disrupts immature and de veloping structures. As suc h, cognitive recovery from the injury is intertwined with contin uing development. Assessing the outcome of injury to a childs developing brain is therefore much like trying to capture a moving target. One must consider not only the relearning of previously acquired skills, as is the cas e with adults, but also the emergence of new skills as the child continues to develop. Thus, it is essential to consider factors such as age at injury, time since injur y, and age at assessment when measuring outcomes after pediatric TBI (Yeates, 2000). Unfortunately, many past studies have failed to take these developmental variables into acc ount. Instead, they often group together all children who have had a TBI within a broad range of ages and m easure them at widely varying times after the injury. For example, a recent longitudinal st udy measured childrens cognitive functioning at three time points: during the fi rst months after admission to th e rehabilitation center, around discharge, and at least 3 y ears after discharge (Van He ugten et al., 2006, pp. 896-897). 12


Although these time points may be clinically relevant markers with respect to recovery, they fail to account for developmental variables such as age at injury and time since injury. A second methodological problem relates to the measurement of attention itself. Previous studies have used dissimilar measures of a ttention, which are usually designed to address different theoretical concepts and are rarely co -normed, complicating comparison across studies. Additionally, the measures themselves are impr ecise, often overlapping with other cognitive domains or failing to control for factors such as motor speed (Fletcher, 1998). For instance, most continuous performance tasks, popularly used to measure sustained attention, rely to varying degrees on working memory and response inhibition. Furthermore, many tasks designed to measure selective attention (e.g., Trail-Ma king-Test [Reitan & Davison, 1974]; Letter Cancellation Test [Talland, 1965]) rely heavily on motor speed, complicating interpretation of scores. Comparing the results of these and other similar attentiona l measures is also problematic, because each test is developed us ing a different normative sample. Another important issue that clouds interpretation of the current literature is the large number of ways in which impairment is determined. Impairment is typically defined by performance at least 1 standard deviation below the mean of some comparison group. Clinically, pediatric neuropsychologists typi cally use an age-matched compar ison group of healthy children. This is also a common practice in the TBI research literature; however, it fails to account for preinjury factors that may be unique to injured child ren, such as inattention. This makes it difficult to conclude that the attentional impairment meas ured after TBI is the co nsequence of the injury itself, rather than a reflection of characteri stics of injury-prone children. There are many possible groups against which a pe diatric TBI sample can be compared: healthy peers, siblings, friends, children with ADHD, and children who have sustained non-TBI traumas. A recent 13


report published by the CDC (Langlois, 2000) strongly recommended that a non-TBI trauma comparison group be used in outcome studies, sin ce they are thought to be the best control for pre-injury factors. However, the use of clinical control group s in the study of attention in pediatric TBI is uncommon, with most studies em ploying healthy, uninjured children or simply comparing the varying severities of pediatric TBI. Despite the methodological challenges and limitations of the existing pediatric TBI literature on attentional impairments, a review of published studies exploring selective attention, sustained attention, divided attention, and attentional control can pr ovide some clues as to the possible pa ttern of deficits we might expect to see within a multidimensional approach. Sustained attention, also known as vigilance, is the ability to maintain the focus of attention over time. In past literature, computerized CPTs utilizing visual stimuli have been used to measure this domain of attention, easing comp arison across studies. Indeed, most of these studies have reported that sustai ned attention is impaired followi ng pediatric TBI, especially for more severely injured children and for children injured at younger ages (Catroppa & Anderson, 2005; Ewing-Cobbs et al., 1998; Fenwick & A nderson, 1999). Wassenberg, Max, and Lindgren (2004) showed that performance on CPTs improv es as time since in jury increases. The performance of children with severe TBIs may be indistinguisha ble from less severely injured children as soon as one year af ter injury (Catroppa & Anderso n, 2005). However, impairments have been reported for severely injured children on more comple x versions of the CPT at two years post-injury, relative to less severely inju red children (Catroppa & Anderson, 2003). Recently, a trend to use auditory measures of sustained attention instead of the CPT has appeared in the literature. Studies using these measures have reported slightly lower performance in children who have incurred severe TBIs compared to healthy controls, an d these differences are 14


present at least five years after injury (Catroppa et al ., 2006; Anderson, Fenwick, Manly, & Robertson, 1998). Clearly, sustai ned attention is impaired foll owing TBI, in both acute and long-term measurements. Divided attention refers to the ability to di vide attention between two or more equally important tasks. It is most commonly measured by dual-task paradigms, in which individuals must attend to two simultaneous and equally im portant events. However, this domain of attention has not been widely stud ied in pediatric TBI, despite th e fact that dual-task paradigms often closely represent r eal-world demands on attention. One of the first studies to investigate divided attention in pediatric TBI found clear evidence of impairments in a sample of children who had sustained moderate to severe TBI an aver age of 6 years prior to testing (Anderson et al., 1998). This study used two dual-tasks: one involv ed a visual selective attention task and an auditory sustained attention tas k, and the other involved an audito ry selective attention task and an auditory sustained attention ta sk. A later study duplicated this result in a similar sample using a similar measure of divided attention (Cat roppa et al., 2006). While the evidence in this area is sparse, there is support for the presence of impairments in divided attention following pediatric TBI several years after injury. Selective attention refers to the ability to focus on relevant stimuli while ignoring irrelevant stimuli. Research on the functioning of selective attention fo llowing pediatric TBI is far from conclusive. Catroppa and Anderson (2005) found that selective attention, as measured by timed tasks involving letter cancellation and connecting a number trail, was impaired in a group of severely injured childre n at one year post-injury. Ho wever, these deficits became insignificant relative to children with mild head injuries by two years post-injury. Likewise, a study by Anderson et al. (1998) re vealed intact selective attention, as measured by timed 15


auditory and visual discrimination tasks, in a sa mple of moderately to severely injured children an average of over 6 years post-injury. A recen t study by Catroppa et al. (2006) further supports this trend. In this study, no signi ficant differences in a motordependent visual discrimination task were found between children who had sustained severe TBI and less severely injured children measured 5 years post-injury. Taken to gether, these studies suggest that selective attention may be impaired in the first year following moderate to severe pediatric TBI, improving to normal levels in subsequent years. However, the tasks used to measure selective attention rely heavily on motor speed, which is known to be inversely associated with injury severity after pediatric TBI (Ja ffe et al., 1993; Yeates, 2000). This suggests that impairments found in selective attention may be partially explained by motor speed impairments. Unfortunately, few studies control for motor speed when measuring selective attention. A study by Ewing-Cobbs et al. (1998) controlled for moto r speed when analyzing selective attention differences between children with mild, moderate, and severe TBIs measured about five to eight years post injury. No differences were found in th ree of the four selectiv e underlining tasks used in the study, although children who had sustained severe TBIs had lower scores on one task involving discriminating between complex shapes. Clearly, more questions about the nature of selective attention following pediatric TBI re main, and differing measures and methodologies have impeded comparison between existing studies. Attentional control, or switc hing/shifting attention, invol ves the ability to change the focus of attention in a flexible and adaptive manner, as well as the inhibition of automatic responses. Clearly, attentional control is a higher level cognitiv e function compared to such domains as selective and sustained attention, but the exact categorization of this cognitive ability is unclear. As such, there are a variety of studies examining this skill, each of which may use a 16


different name or label to refer to this func tion. For example, researchers who focus on higher level cognitive functions often refe r to attentional control as an executive function, whereas those who study attention often include it as a domain of attention. Thus, attentional control has been studied in multiple ways, in both executive functi ons literature and attention literature. Since there are many ways in which attentional control has been conceptualized, there are also multiple ways of measuring it. Tasks such as the Wisc onsin Card Sorting Task (WCST; Heaton, Chelune, Talley, Kay, & Curtiss, 1993) are popular in the executive f unctions literature, while the attention literature tends to focus on various re sponse inhibition tasks such as the Contingency Naming Test (CNT; Taylor, Schatsneider, & Ri ch, 2002). Despite the various ways that attentional control has been approached in the literature, most studies have shown that it is impaired following pediatric TBI. Impaired pe rformance on the WCST has been documented in the first few years following injury in severely in jured children, but these deficits seem to fade by about 5-8 years following injury (Van Heugten et al., 2006; Ewing-Cobbs et al., 1998). It has also been shown that impairments in atten tional control, as measured by timed response inhibition tasks, are related to injury severity (Catroppa et al., 2006). However, whether observed deficits in attentiona l control tasks may actually be caused by slowed processing speed remains to be determined (C atroppa & Anderson, 2005). Given the lack of consensus in studies to date, more studies are needed to help characterize the pattern of attentional impairments following seve re pediatric TBI. The current study seeks to unify past research by using a multidimensional model of attention and using co-normed measures representing four popular attentional do mains. Furthermore, the current study will utilize an orthopedic control group to account for pre-injury risk factors for injury, which has been a neglected practice in past literature. Th e purpose of the current study is to characterize 17


the pattern of attentional skills w ithin 1 year afte r moderate to severe TB I in a group of children and adolescents who participated in a longitudinal research study examining the effects of childhood head injury on memory and attention. Based on th e previous research, it was hypothesized that sustained and divided attentio n would be impaired re lative to the control group. In light of prior findings of deficits in attentional control, measured both as an executive function and as an attentional domain, it was hypothes ized that attentional control in the current study would be impaired as well. In contra st, selective attention was hypothesized to be unimpaired after controlling for motor speed, cons idering the heavy reliance of most selective attention tasks on motor speed, and evidence in the li terature of spared selective attention skills. 18


CHAPTER 2 METHODS Participants In total, forty-four children participated in the present study. The sample was comprised of 26 children who had sustained a moderate ( n=3) or severe ( n =23) traumatic brain injury within the past year (TBI group) and 18 children who had sustained an orthope dic injury not involving the head (OI group). Inclusion criteria for a ll participants were as follows: (1) documented history of moderate or severe TBI within the pa st year or documented history of orthopedic injury with no injury to the head or brain within the past year; (2) age at injury between 6 and 16 years; and (3) English-speaking. Exclusion criteria were as follo ws: (1) pre-injury history of neurological or developmental diso rder; (2) previous head injury; and (3) pre-injury diagnosis of an attentional disorder. Moderate to severe TBI was defined as a Glasgow Coma Scale (GCS; Teasdale & Jennett, 1974) score 12 upon arrival at the hospital, loss of consciousness (LOC) > 1 hour, and presence of brain and/or skull abnormalities on CT or MRI scan. Mechanisms of injury in the TBI group included motor vehicle accidents ( n=17), recreational vehicle vs. motor vehicle accidents (n =4), and falls ( n=5). The OI group had somewhat different injury mechanisms, including falls (n=8), sports-related injuries ( n=8), and one recreational vehicle roll-over. One childs mechanism of injury was unclear, but resulted in an orthope dic surgery of the knee. Table 2-1 shows demographic information for each group. The TBI and OI groups did not differ from one another with respect to age at injury, age at testing, gender, or ethnicity. There was also no statistically si gnificant difference detected betw een the groups with respect to family socioeconomic status ( t (29.24) = -0.89, p= .378), as measured by the two-factor 19


Hollingshead Index (Hollingshead, 1957) which is calculated based on pa rental education and occupation levels. Procedures Children were recruited between Januar y 2002 and April 2007. The TBI group was recruited from the University of Florida and Sha nds Hospital in Gainesville, Florida, while the OI group was primarily recruited via community physic ian referrals and flyers in the same city. After the purpose of the study was explained and the children were determined to me et eligibility criteria, interested parents signed a consent form and scheduled a visit. Before any testing took place, each child gave signed or verbal assent to participate. The children were recruited as part of a larger longitudinal study examining attention and memory following pediatric head injury, and they were paid $20-40 for completion of each vi sit. However, for the purposes of this study, only data from the first assessment within one ye ar following injury was analyzed. This study was approved by the Institutional Review Board of the Health Science Center at the University of Florida, Gainesville. Medical information needed to determine se verity of injury in the TBI group (e.g. GCS score, duration of LOC) was collected from each childs medical record. The neuropsychological data analyzed in this study were collected dur ing the second hour of a larger test battery, following two abbreviated meas ures of intelligence and achievement. Measures Parent Report Measures Basic participant information Parents completed a basic study questionnaire that provided demographic and injury-re lated information, as well as information about their childs psychological and medical history. 20


Pre-injury functioning. Although children with formally diagnosed attentional disorders (i.e., ADHD) were excluded from the present study, pre-injury atte ntional functioning was assessed to ensure that the results could not be at tributed to systematic gr oup differences in this area. The Conners Parent Rating Scale-Revise d: Long Version (CPRS-R:L; Conners, 1997) is a reliable and valid 80-item questi onnaire widely used to assess at tention and behavior problems in children. Parents answer questions about their childs beha vior over the past month on a fourpoint Likert scale (ranging from Not at all to Very much) and scores are produced for 14 subscales (e.g., Anxious-Shy, Oppositional Behavior). In this study, data from the Cognitive Pr oblems/Inattention, DSM-IV Inattention, and ADHD Index subscales were analyzed as estimates of each childs pre-injury attentional functioning. High scores ( t 65) on the Cognitive Problems/Inattention subscale suggest troubles with organization, concentration, and completing tasks. High scores on the ADHD Index suggest children who ar e at risk for developing ADHD, and high scores on the DSM-IV Inattention scale corre spond to the DSM-IV criteria fo r ADHD-Inattentive type (e.g., making careless mistakes, easily distracted). By nature of the study design, parents were asked to rate thei r childs functioning retrospectively. To maximize the validity of these ratings, the CPRS-R:L was completed as soon as possible following recruitment ( m =90.0 days; SD =86.9 days). However, there are several limitations to this method. Parents may simply forg et how a child acted before his or her injury, or the childs post-injury functi oning may affect parents reports Furthermore, having come close to losing a child, parents may have a te ndency to underreport pre-injury problematic behaviors due to a change in perspective or priorities. Unfortunately, inaccuracies in 21


retrospectively reported information are inevitable; thus, these ratings are best interpreted as estimates of pre-injury functioning. Child Assessment Attentional functioning Attentional abilities were eval uated using the Test of Everyday Attention for Children (TEA-Ch; Manly, Anderson, Robertson, & Nimmo-Smith, 1999), a standardized test that uses ni ne game-like subtests to measure several domains of attention: selective, sustained, divided, a nd attentional control. The TEA -Ch has been shown to have acceptable test-retest reliability and convergence va lidity with other tests of attention, and it is sensitive to attentional impairments seen in both Attention Deficit/Hyperactivity Disorder (ADHD) and TBI. Three domains (selective, su stained, and attentional control) have been supported by structured equation models; thus, scores from the TEA-Ch subtests combine to form three composite scores. Alth ough a factor analysis designated the dual task subtests of the TEA-Ch as part of the sustained attention compos ite, past attentional research has used similar tasks to measure divided attention. As such, one of these tasks was sele cted to represent the divided attention component in the present study. These subtests are co-normed, allowing for comparison across domains, and scaled scores take both age and gender into account. For this study, performance on four of the nine subtests was analyzed in order to d ecrease the number of variables while still sampling ea ch of the attentional domains. Sustained attention Score! is a 10-trial auditory sust ained attention task that involves silently counting scoring tones played on an a udio tape. Tone counts for each item range from 9-15 and occur at varying intervals. The score of this subtest is based on the number of trials for which the child provided th e correct tone count. Divided attention. Score! DT is a 10-trial divided a uditory attention task, which involves silently counting tones (as required in the Score! subtest) while at the same time listening out for 22


the name of an animal mentione d during a mock news broadcast. Children must state both a tone count and an animal name following each trial, and the score is based on the total number of animals plus the total number of tone counts correct (maximum score = 20). This subtest was used because it does not require motor out put; thus, motor speed does not complicate interpretation of the scores. Selective attention The Sky Search task of the TE A-Ch involves visual scanning and discrimination. In this task, children are timed as they circle target figures which are located within an array of similar-looking distractor figures. Th ere is also a motor control component in which the child must circle target figures in the absence of distractors. The Sky Search subtest yields three scores: num ber of targets found, time per target and an attention score, which represents time-per-target after controlling fo r motor speed. This subtest was chosen in particular for its ability to c ontrol for motor speed, a factor that is commonly overlooked in measures of selective attention. Attentional control Creature Counting is a 7-trial timed task that involves switching between counting forwards and backwards while counting creatures in a winding row. Throughout the row, arrows signal the child to count forwards (upward pointing arrow) or backwards (downward pointing arrow) from the last counted creature The trial is considered to be correct if the child obtains the correct count at the end of the row, which suggests that the child was able to switch between counting fo rwards and backwards properly upon encountering each arrow. Number of counting switches required to obtain a correct end count ranges from 2 to 6. This subtest provides two scores: 1) number of correct trials and 2) time-per-correctswitches (a measure of accuracy). Time per-correct-switches is not calculated if the total correct 23


trials is two or less. For the purposes of this study, child ren who fell in this range ( n =5) were assigned a scaled score of 1 for time-per-correct-switches. Statistical Analysis The TBI and OI groups were first compared on demographic and preinjury variables to determine whether there were significant differen ces prior to injury. TEA-Ch subtest scores were converted into standardized scaled scores using the published normative data, which adjusts for both age and gender. Data was screened for outliers as well as violations of normality and equal variance assumptions. Independent sample s T-tests with Bonferr onis correction were used to compare the TBI and OI group mean scaled scores on the four TEA-Ch subtests. Results were considered statistically significant if p<0.0125. Levenes test for equality of variances was significant for the Cognitive Prob lems/Inattention scale of th e CPRS-R:L and the Creature Counting total correct score of the TEA-Ch; p-values were adjusted accordingly. Furthermore, the attentional performance of the TBI group was analyzed descriptively to provide clinically relevant information about impairments in head-i njured children relative to a normative sample of healthy, age-matched children. Despite the fact that this method of anal ysis does not control for pre-injury factors, it provides a full picture of the attentional difficulties that will need to be addressed in a clinical assessmen t of childhood TBI. For this de scriptive analysis, group mean scaled scores of 7.0 or less (i.e. 1 SD below the scale mean) were considered to represent clinically significant impairment. This cut-off re presents performance at or below one standard deviation unit below the mean of the normative sample, a frequently employed cut-off for clinical neuropsychological evaluations. 24


Table 2-1. Participant demographics TBI group OI group Variable ( n=26) ( n=18) Age at testing, mean months ( SD ) 154.08 (43.06) 141.44 (37.87) Age at injury, mean months (SD ) 151.69 (45.47) 138.60 (37.60) Time since injury, mean days (SD) 102.69 (95.76) 72.80 (86..40) Gender, percent male 50 61.1 Ethnicity, percent Caucasian 69.2 72.2 Hollingshead Index 1, 2 mean score ( SD ) 30.15 (41.56) 22.56 (10.23) 1 Hollingshead Index was unable to be calcu lated for two children in the TBI group. 2 Equal variances not assumed. 25


CHAPTER 3 RESULTS Data screening revealed no outliers and no vi olations of the assumption of normality. However, four measures (Hollingshead Index; CPRS-R:L Cognitive Problems/Inattention; TEACh Score! DT, Total correct items; TEA-Ch Creat ure Counting, Total correct ) violated the equal variances assumption, and adjusted degrees of freedom were used in analyses of these measures (reported below). Pre-Injury Functioning Table 3-1 shows the mean T scores for each group on the selected subscales of the CPRSR:L. The orthopedic injury (OI) group was rated significantly higher on both pre-injury Cognitive Problems/Inattention ( t (24.68)=2.61, p =0.015) and DSM-IV Inattentive symptoms ( t (40)=2.63, p=0.012). Furthermore, there was a tre nd for the OI group to have a higher ADHD Index. Although this difference was not statistically significant, there was a medium effect size of group (Cohens d=-.054). Despite these differences in pa rental rating of preinjury attentional functioning between the OI and TBI groups, neither groups mean T scores we re in the clinical range as defined by the CPRS-R:L manual ( t 65). As such, it is unlikely that the higher estimate of pre-existing attentional diffi culties in the OI group was of a large enough magnitude to influence the results. Nevertheless, this poten tial confound was addressed in post hoc testing. The post hoc analysis revealed that the pattern of results, described below, did not change after removal of children ( n=5) with clinically significant elevat ions on any of the CPRS-R:L scales. Attention Measures Sustained Attention Analysis of the Score! task revealed that on average, the TBI group scored nearly 3 scaled score points lower (i.e., 1 SD) than the OI group. This difference was st atistically significant 26


( t (41)=2.73, p=0.009), and the effect size was large (Cohens d=-0.85). Furthermore, the descriptive analysis show ed that the TBI groups mean scaled score of 7.31 ( SD =3.62) hovered just outside the clinically signi ficant range, with 15 of the 26 ch ildren (57.7%) performing at or below the cutoff. In contrast, the OI group mean ( m =10.22, SD =3.28) was well within normal limits, with only 5 of the 18 child ren (27.8%) performing at or below the cutoff. These results reveal that the TBI groups performance in sust ained attention was significantly lower than the OI group and quite close to meeting the cu toff for clinically significant impairment. Divided Attention Analysis of the Score! DT task revealed that children in the TBI group had significantly worse performance on this measure of divided auditory attention ( t (40.32)=3.13, p=0.003). The mean scaled score of the TBI group was 7.60 (SD =4.31), over 3 scaled score points (i.e., >1 SD) lower than the OI group ( m =10.94, SD =2.69), and this difference is reflected in a large effect size (Cohens d=-0.92). Although the TBI groups low pe rformance on the Score! DT task provides evidence of a st atistically significant difference from the OI group, it did not fall within the clinically significant range. Selective Attention Table 3-2 presents the mean scaled scores for each group on the Sky Search task. There were no statistically significant di fferences in mean scaled scores between the TBI and the OI groups on the Sky Search task in terms of total targets, time-per-target, or Attention score. While not statistically significant, ther e was a medium effect size (Cohens d=-0.58) for group differences on time-per-target, indi cating that there was a trend for the TBI group to be slower on the task despite finding a similar number of targets compared to the OI group ( t (41)=1.11; p=0.274). The effect size for the Sky Search At tention score, which controls for motor speed, 27


was much smaller (Cohens d=-0.25), supporting the in terpretation that differences in selective attention may be at least partially ex plained by differences in motor speed. Although there were no statisti cally significant differences between groups on the Sky Search task, the TBI groups mean scaled score of 6.2 ( SD =3.66) on time-per-target represents a clinically significant impairment. In total, 15 of the 25 children (60%) in the TBI group who completed this task performed at or below th e cutoff. Furthermore, the TBI groups mean Attention score of 7.32 ( SD =4.01), is just outside the clinica lly significant range, and 10 of the 25 children (40%) in the TBI group performed in the clinically si gnificant range. Thus, there is evidence for clinically relevant impairment in both motor speed and selective attention. However, the TBI group was able to identify an average number of targets ( m =10.2, SD =3.04) when time was not taken into account at all. Attentional Control Table 3-2 presents mean scaled scores for both groups on the Creature Counting task. The TBI and OI groups did not differ with respect to total correct trials ( t (40.78)=-0.38, p=0.707). Furthermore, both groups had mean scaled scores for total correct that were within the average range (TBI m =9.46; OI m =9.12). However, the TBI group ha d significantly worse performance on the Timing score (time-per-correct-swi tches) relative to the OI group ( t (41)=2.67, p=0.011), and their mean scaled score of 5.0 ( SD =3.68) represents clinically significant impairment. In total, 20 of the 26 children in th e TBI group (77%) performed at or below the cutoff for clinical significance. This suggests th at the TBI group took a significan tly longer amount of time to switch between counting forwards and backwards, even though they were unimpaired in their ability to perform the task correctly. Thus, th eir impairment may be a reflection of impaired processing speed relative to the OI group and to the healthy ageand gender-matched normative sample. 28


Group means for the attentional tasks are also represented visually in Figure 3-1. 29


Table 3-1. Parent ratings of pre-injury a ttentional performance (T-scores) TBI group OI group t -value p-value Cohen's ( n=25) 1 ( n=17) 1 d Cognitive Problems/Inattention 46.84 (5.98) 53.59 (9.45) 2.61 0.015 2 -0.91 ADHD Index 47.40 (7.17) 51.12 (6.87) 1.68 0.101 -0.54 DSM-IV Inattentive 46.80 (5.78) 52.06 (7.15) 2.63 0.012 -0.85 Table 3-2. Performance of study groups on the TEA-Ch (scaled scores) TBI group OI group t-value p-value Cohens ( n=26) ( n=18) d Sustained Attention Score! : Total correct 7.31 (3.62) 10.22 (3.28) 2.73 0.009 -0.85 Divided Attention Score! DT: Total correct items 7.60 (4.31) 2 10.94 (2.69) 3.13 0.003 4 -0.92 Selective Attention Sky Search: Total targets 10.20 (3.04) 1 11.17 (2.48) 1.11 0.274 -0.04 Sky Search: Time-per-target 6.20 (3.66) 1 8.06 (2.65) 1.83 0.074 -0.58 Sky Search: Attention score 7.32 (4.01) 1 8.22 (3.19) 0.79 0.434 -0.25 Attentional Control Creature Counting: Total correct 9.46 (3.51) 9.12 (2.45) 3 -0.38 0.707 4 0.11 Creature Counting: Timing score 5.00 (3.68) 7.94 (3.34) 3 2.67 0.011 -0.85 Note: SD appears in parentheses. 1 n=25; 2 n=25; 3 n=17; 4 Equal variances not assumed. 30


Figure 3-1. Attentional performan ce of children with moderate-severe TBI or orthopedic injuries 31


CHAPTER 4 DISCUSSION Attentional testing within the first year fo llowing moderate to severe pediatric TBI revealed a pattern of impairment across multiple domains of attention. Consistent with our hypotheses, children who had sustained moderate to severe pediatric TBI exhibited statistically significant deficits in sustained attention, divided attention, and a ttentional control. Clinically significant impairment was found in the domain of attentional control, and near-clinically significant impairment was found in sustained atte ntion. However, the TBI groups performance in divided attention was not below the clinical cutoff for im pairment. The hypothesis that selective attention would be unimpaired only after controlling for motor speed was not supported. Although deficits in se lective attention were not appa rent in comparison to the OI group after controlling for motor speed, there we re no significant diffe rences between groups before controlling for motor speed. Despite a lack of the predicted group differences in speeded performance, the TBI groups performance in this area was suggestive of c linical impairment in comparison to the normative sample. Thus it appears that there is evidence for attentional impairment across all four domains of attenti on examined following moderate to severe TBI. The sustained attention deficits found in th e present study are consistent with previous studies, which have shown that children have diffi culties in maintaining attention over time after moderate to severe TBI (Ewi ng-Cobbs et al., 1998; Catroppa & Anderson, 2005; Wassenberg et al., 2004). In these studies, sustained atten tion was measured via a visual continuous performance test. However, the present study found evidence for impairments in selective attention measured via an auditory task, providing evidence that auditory measures of sustained attention are sensitive to deficits in sustained at tention after pediatric TBI. This is consistent with evidence from two othe r studies (Catroppa et. al., 2006; Anderson et al., 1998). 32


As mentioned above, the present study had mixed findings regarding divided attention performance. The TBI group was significantly imp aired relative to the OI group, but their mean scaled score ( m =7.60) was in the average range relative to the normative sample. This suggests that while TBI group had poorer performance than would be expected, it is not outside the normal range of attentional skills. This finding is inconsistent with pa st studies, which have found impaired divided attenti on in children who have sustai ned TBIs in comparison to uninjured children (Catroppa et al., 2006; Anderson et al., 1998). Of note, the TBI group in the present study had a large standard deviation on the divided attention task ( SD =4.31). It is possible that the mean score was raised by a sm all number of high performers in the TBI group, thus masking the difficulties that the majority of the group had on the task. Indeed, nearly half of the TBI group (11 of 25) performed at or below the cutoff for clinical impairment. Still, the question of whether or not divide d attention is impaired follow ing pediatric TBI needs to be explored further in futu re studies. Looking closer at the pattern of performance on the selective attenti on task offers some explanations for why our hypothesi s regarding this domain was not fully supported. It was noted that the TBI group located an average number of targets in co mparison to both the OI group and the normative sample. This suggests that modera te to severe TBI does not interfere with a childs ability to accurately discriminate between target stimuli and distractors. However, the TBI group performed this task slowly, resulting in a statistically and clinically significant impairment in time-per-target. Past literature ha s shown that motor speed is impaired in children following TBI, especially for children with more severe injuries (Yeat es, 2000; Jaffe et al., 1993). One study identified specific deficits in upper limb speed and dexterity in a sample of children with TBIs measured at least 16 months after injury (Chaplin, Deitz, & Jaffe, 1993). 33


Upper limb speed and dexterity are large compone nts of the motor abili ties required by the Sky Search task used in the presen t study. Controlling for motor sp eed, as is possible due to the design of the TEA-Ch task, allowed for a modest improvement of about one standard score point, but the resulting m ean Attention score was still below th e clinical cutoff for impairment. Clearly, motor speed impairments do not fully explai n the deficit in selective attention seen after pediatric TBI. Furthermore, this deficit cannot be explained by difficulty discriminating between target and distractor stimuli, since no im pairments in this ability were seen. A similar pattern was seen for Creature Countin g, the task designed to measure attentional control. The TBI group had an average number of correct trials but an impaired timing score in comparison to both the OI group and the normative sa mple. This suggests that the children did not have difficulty accurately switching between the tasks; rather, making these switches efficiently was the problem. Given that this ta sk has only a small ocular motor component, this time-related impairment cannot be attributed to motor speed. Instead, it se ems that mental speed, commonly referred to as processing speed, may contribute strongly to this low score. Processing speed, generally defined as the rate at which cognitive processes occur, has not been fully explored in the pedi atric TBI literature. However, evidence from adult TBI studies may offer insight into how processing speed is affected by TBI. A recent meta-analysis of 41 studies showed that adults who have sustaine d severe TBIs typically perform one standard deviation below normal controls on both simple and complex measures of processing speed (Mathias & Wheaton, 2007). Additiona l studies of adult TBI have al so suggested that processing speed slows as the complexity of a task incr eases (Tombaugh, Rees, Stormer, Harrison, & Smith, 2007). A pattern of average accuracy despite long er time to completion, as found in the present 34


study, has been documented in adult TBI populations on tasks of attentiona l control (Ponsford & Kinsella, 1992). Studies of processing speed following pediatri c TBI are less conclusive. Slowed reaction times on computerized continuous performance task s (CPTs) have not been found after pediatric TBI (Catroppa & Anderson, 1999; Robin, Max, Stierwalt, Guenzer & Lindgren, 1999; Catroppa & Anderson, 2005), suggesting that processing speed is unimpaired on this type of task. In contrast, some studies have found low performa nce on the Processing Speed Index (PSI) of the Wechsler Intelligence Scale for Children, Th ird Edition (WISC-III) after pediatric TBI (Calhoun & Dickerson Mayes, 2005; Catropp a & Anderson, 2005). However, it is important to note that motor speed is not measured or accounted for on th e tests that comprise the PSI, despite the fact that the tests involve motor output. Furthermore, these tests require the child to discriminate between target and distractor s timuli, making it difficult to dis tinguish between processing speed and selective attention. The overlap between measures of processing sp eed and selective atte ntion has also been addressed in the litera ture exploring Attenti on Deficit/Hyperactivity Disorder (ADHD). One notable study in this area suggest ed that a processing speed defic it is the basis for symptoms of inattention seen in children diagnosed w ith the Inattentive su btype of Attention Deficit/Hyperactivity Disorder (ADHD; Weiler, Holmes Bernstein, Bellinger, & Waber, 2000). In this study, children with ADHD -Inattentive type had normal motor speed but impaired scores on the PSI. This suggests that cognitive pr ocessing speed impairments and symptoms of inattention (e.g. being easily di stracted, making careless mistak es) are related; however, the authors were careful to note that a clear distinctionbe tween the constructs of processing speed and selective attention is often difficult to make (Weiler et al., p. 228). 35


In the present study, the pattern of average a ccuracy in the context of impaired speeded performance (even after contro lling for motor speed) suggests that deficits in cognitive processing speed may, indeed, be a plausible expl anation for the speed-related impairment seen in both attentional contro l and selective attention. Unfort unately, a pure measure of processing speed was not obtained in the present study, ha mpering any further exploration of this hypothesis. Nevertheless, there is a strong possibility that proce ssing speed deficits contributed to the pattern of results. This brings to lig ht the importance of taking both motor speed and processing speed into account when measuring a ttention in brain-injure d populations. In fact, one could argue that the attenti onal deficits seen after pediatri c TBI are secondary to motor and processing speed impairments, just as it has b een suggested that processing speed impairments are the basis of attentional de ficits seen in ADHD-Inattentive type (Weiler et al., 2000). However, the extent to which the impairments seen in this study are reflective of primary deficits in attention is largely irrelevant when it comes to a childs functio ning outside the laboratory. In everyday life, a child must perform with accuracy and efficiency in order to succeed. Put simply, speed is always a factor. Primary and secondary deficits in attention are likely to have a similar negative impact on a childs academic, social, and behavioral functioning, and any discussion of the distinction between the two should not lose sight of this fact. Nevertheless, the distinction between primary and secondary deficits in attention is important in the development of accommodations and rehabilitative strate gies. If a childs attentional difficulties are sec ondary to motor and processing speed impairments, then the standard recommendation for increased time allo wances on school tests and assignments is a helpful one. However, if a child has a primary atte ntion deficit, then increased time will be of no 36


benefit. Impaired accuracy on tasks like Sky Search and Creature Counting, which would be suggestive of a primary deficit in attention, may be predictive of continued poor academic functioning after implementation of academic acco mmodations that involve extended time. On the other hand, average accuracy but impaired efficiency on these tasks, as was found in the present study, would be suggestive of a secondary deficit in attention. This pattern may be predictive of improved academic functioning upo n allowance of extended time for tests and assignments. The distinction betw een primary and secondary deficits in attention after pediatric TBI and its potential implicati ons for the prediction of response to academic accommodations is an area that deserves further study. It has been proposed that controlling for pr ocessing speed may leave surprisingly little evidence for impairments in selective, divided, and sustained attention after adult TBI (Zomeren & Brouwer, 1994, p. 90). As discussed above, ev idence from the present study supports this statement in the domain of sel ective attention. Questions rema in, however, about the role of processing speed in sustained and divided atten tion. In the present st udy, children in the TBI group performed poorly on the sustained and divided attention tasks in comparison to the OI control group. Interestingly, th ese tasks have externally impos ed time restraints, preventing children from slowing down to compensate for a ny processing speed deficits. In contrast, children were able to control the length of time spent on the selective attention and attentional control tasks, and their accuracy was in the aver age range. While this may be explained simply as a differential pattern of performance across attentional doma ins, it could also reflect a negative impact of external time control on performance. Allowi ng children to control the length of time spent on a task may enable them to compensate for slow processing speed. If external time control does indeed impair performance, it is possible that allowing children more time on 37


tasks of sustained and divided attention (e.g. by increasing inter-stimul us intervals) would improve accuracy. Direct testing of this hypothesis was not possible within the design of the current study, but it is an interesting area for future study. Although this study suggests that attentional functions and pr ocessing speed are somewhat separate constructs, some theories of attention po sit that attention is a component of information processing (see Cooley & Morris, 1990 for a brief re view). In these theori es, attention functions as a filter to narrow down the amount of informati on that is to be proces sed. When attention is intact, irrelevant information is filtered out, allowing proces sing to take place efficiently. However, if the attentional filter is damaged in some way, the brains processing networks are overwhelmed with information, slowing down the speed with which it can be processed. Viewing attention via this type of theory suggest s that slowed processing speed may not cause a secondary impairment in attention; rather, impa ired attention may cause a secondary impairment in processing speed. A clearer picture of exactly what the construct of attention is and how it functions in the brain will deve lop as future studies investigat e the overlap between processing speed and attentional functions. While differences in attentional ability, processing speed, and motor speed may explain much of the variation between groups in the present study, it is possible that other factors contribute as well. One possibility is that a ttentional performance is under some degree of voluntary control. If this were the case, subjec ting children to time pressure, rather than allowing them to proceed at their own pace, may improve their performance in some domains. This possibility was explored in post hoc testing. Only four of the nine subtes ts of the TEA-Ch were analyzed in the present study, in an effort to decrease the number of dependent va riables and preserve power. However, a fifth 38


subtest, Map Mission, was analyzed post hoc in order to explore the effect of external time pressure on attentional performance. Map Miss ion is a test designed to measure selective attention. The child is presented with a large map, over which an array of target and distractor map symbols (e.g., gas pump, knife a nd fork) are superimposed. The child is required to circle as many target symbols map as possible in sixt y seconds. Although this task does not provide a control for motor speed, it provides an interestin g comparison for the Sky Search task. On the Sky Search task, children are able to proceed at their own paces, while the Map Mission task requires children to perf orm within an explicitly stated tim e limit of 1 minute. Interestingly, children who had sustained a TBI did not perfor m significantly below the OI group on the Map Mission task ( t (40.99) = 1.62, p=0.11), and their group mean performance was in the average range relative to the normative sample ( m =8.52; SD =4.59). In comparison, the primary study findings indicated that the TBI gr oup was impaired on the Sky Search task even after controlling for motor speed. As such, it seems that childr en in the TBI group had improved performance in selective attention when given specific time limits. It seems that when they were allowed to take more time, they do, but when they were required to perform within strict time limits they were still able to achieve av erage performance. Of course, this comparison is not straightfo rward. There is a notable difference between the Sky Search and Map Mission tasks: Sky Search targets are larger but require more complex processing, whereas Map Mission targ ets are smaller, but simpler. It is possible that the poor performance of the TBI group on the Sky Search ta sk could be attributed to the more complex nature of the stimuli. Nevertheless, it is interesting that the TBI group was unimpaired on the Map Mission test of se lective attention, even without c ontrolling for motor speed. The possibility that children are able to achieve pe rformance in the average range when they must 39


function within strict time limits has interes ting implications for academic accommodations. Perhaps merely stressing the importance of effici ent performance, rather than allowing extended time, is enough to accommodate for deficits seen afte r pediatric TBI, at least on some tasks. The relationship between time pressu re and performance after TBI is an important area for future research. Our study has several strengths. It approaches attention as a multidim ensional construct. The tests used to measure each attentional domain are co-normed, allowing for a comparison between the tests and a discussion of the patter n of functioning across them. Also, it makes use of an orthopedic injury control gr oup, allowing pre-injury characteris tics to be controlled for. Despite these strengths, the st udy does have some limitations. The sample size was relatively small; however, the large effect sizes indicate th at significant differences were able to be detected statistically when they were present. The socioeconomic status differences noted between the two groups may also be considered as a limitation, although no association between SES and attentional performance was found. Un fortunately, there was limited information available about the extent and lo cation of the brain injuries, whic h prevented any analysis of how attentional domains may be differentially imp acted by injury to different neuroanatomical regions. Mirsky et al. (1991) suggested that abilit ies in the different attentional domains may be served by distinct cortical regions, which gives rise to the possibility that location of injury may play a large role in the pattern of attentional functioning after pediatric TBI. Also, the wide variation in time since injury may have prevente d the discovery of a meaningful pattern of postinjury attentional performance. A final limita tion is the fact that children with pre-existing attentional disorders (e.g. ADHD) were exclude d from the analysis. This may impact the 40


generalizability of th e study, since ADHD is over-represented in the population of children who sustain TBIs (Gerring et al., 1998). These limitations could be addressed by future studies. In summary, our study found evidence for impa irment in selective, sustained, and divided attention, as well as attentional control after modera te to severe pediatric TBI. It also raised questions about the role that motor speed and pr ocessing speed play in these deficits. Future studies could also explore the ques tions that have been left unanswered by the current study. In addition to exploring the primary vs. secondary nature of attentional difficulties after pediatric TBI and the impact of time pressure on attentio nal performance, future studies could explore how the pattern of attentional impairment cha nges as a child recovers from brain injury. 41


LIST OF REFERENCES Anderson, V., Fenwick, T., Manly, T., & Robert son, I. (1998). Attentional skills following traumatic brain injury in ch ildhood: A componential analysis. Brain Injury 12(11), 937949. Biederman, J., Monuteaux, M.C., Mick, E., Spencer, T., Wilens, T.E., Klein, K., et al. (2006). Psychopathology in females with attention-de ficit/hyperactivity diso rder: A controlled, five-year prospective study. Society of Biological Psychiatry, 60, 1098-1105. Calhoun, S.L., & Dickerson Mayes, S. (2005). Pro cessing speed in children with clinical disorders. Psychology in the Schools, 42 (4), 333-343. Catroppa, C., & Anderson, V. (1999). Attentional skills in the acute phase following pediatric traumatic brain injury. Child Neuropsychology, 5 (4), 251-264. Catroppa, C., & Anderson, V. (2003). Children's attentional skills 2 years post-traumatic brain injury. Developmental Neuropsychology, 23 (3), 359-373. Catroppa, C., & Anderson, V. (2005). A prospectiv e study of the recovery of attention from acute to 2 years following pedi atric traumatic brain injury. Journal of the International Neuropsychological Society, 11 (1), 84-98. Catroppa, C., Anderson, V., Morse, S., Haritou, F., & Rosenfeld, J. (2006). Childrens attentional skills 5 years post-TBI. Journal of Pediatric Psychology, 32 (3), 354-369. Chaplin, D., Deitz, J., & Jaffe, K.M. (1993). Motor performance in children after traumatic brain injury. Archives of Physical Medicine and Rehabilitation, 74 (2), 161-4. Conners, C.K. (1997). Conners rating scales-revised. New York: Multi-Health Systems. Cooley, E. J., & Morris, R. D. (1990). Attent ion in children: A Neur opsychologically based model for assessment. Developmental Neuropsychology, 6 (3), 239-274. Dennis, M., Wilkinson, M., Koski, L, & Humphreys, R.P. (1995). Attention deficits in the long term after childhood head injury. In S. H. Broman & M.E. Michel (Eds.), Traumatic head injury in children (pp.165-187). New York: Oxford University Press. Ewing-Cobbs, L., Barnes, M., Fletcher, J.M., Levin, H.S., Swank, P.R, & Song, J. (2004). Modeling of longitudinal academic achievement scores after pediat ric traumatic brain injury. Developmental Neuropsychology 25, 107-133. Ewing-Cobbs, L., Prasad, M., Fletcher, J.M., Levin, H.S., Miner, M.E., Eisenberg, H.M. (1998). Attention after pediatric traumatic brain injury: A multidimensional assessment. Child Neuropsychology, 4 (1), 35-48. 42


Fenwick, T., & Anderson, V. (1999). Impairme nts of attention following childhood traumatic brain injury. Child Neuropsychology, 5 (4), 213-223. Fletcher, J.M. (1998). Attent ion in children: Conceptual and methodological issues. Child Neuropsychology, 4 (1), 81-86. Gerring, J.P., Brady, K.D., Chen, A., Vasa, R., Gra dos, M., Bandeen-Roche, K. J., et al. (1998). Premorbid presence of ADHD and development of secondary ADHD after closed head injury. Journal of the American Academy of Child and Adolescent Psychiatry, 37 (6), 647-654. Heaton, R.K., Chelune, G.J., Talley, J.L ., Kay, G.G., & Curtiss, G. (1993) Wisconsin Card Sorting Test manual: Revised and updated. Odessa, FL: Psychological Assessment Resources. Hollingshead, A. (1957). Two factor index of social position. New Haven, CT: Yale University. Hooper, S.R., Alexander, J., Moore, D., Sasher, H ., Laurent, S., King, J., et al. (2004). Caregiver reports of common symptoms in children following a traumatic brain injury. NeuroRehabilitation, 19, 175-189. Jaffe, K.M, Fay, G.C., Polissar, N.L., Martin, K.M ., Shurtleff, H.A., Rivara, J.M., et al. (1993). Severity of pediatric traumatic brain injury and ne urobehavioral recove ry at one year--a cohort study. Archives of Physical Me dicine and Rehabilitation, 74 (6), 587-595. Keenan, H.T. & Bratton, S.L. (2006). Epidemiology and outcomes of pediat ric traumatic brain injury. Developmental Neuroscience, 28, 256-263. Kraus JF. (1995) Epidemiological features of brain injury in children: oc currence, children at risk, causes and manner of in jury, severity, and outcomes. In S.H. Broman & M.E. Michel. (Eds.), Traumatic head injury in children (pp. 22-39). New York: Oxford University Press. Langlois, J.A. (Ed.). (2000). Traumatic brain injury in the United States: Assessing outcomes in children. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Langlois, J.A., Rutland-Brown, W., & Thomas, K.E. (2006). Traumatic brain injury in the United States: Emergency department visits, hospitalizations, and deaths. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Lescohier, I., & DiScala, C. (1993). Blunt trauma in children: Causes and outcomes of head versus extracranial injury. Pediatrics, 91 (4), 721-725. 43


Levin, H.S., Mendelsohn, D., Lily, M.A., Yeakley, J., Song, J., Scheibell, R.S., et al. (1997). Magnetic resonance imaging in relation to f unctional outcome of pediatric closed head injury: A test of the Ommaya-Gennarelli model. Neurosurgery, 40(3), 432-441. Manly, T., Anderson, V., Robertson, I., & Nimmo-Smith, I. (1999). The test of everyday attention for children. London, England: Thames Valley Test Company. Mathias, J.L., & Wheaton, P. (2007). Changes in attention and informati on-processing speed following severe traumatic brain injury: a meta-analytic review. Mirsky, A.F., Anthony, B., Duncan, C.C., Ahearn, M ., & Kellam, S.G. (1991). Analysis of the elements of attention: A neuropsychological approach. Neuropsychology Review, 2 (2), 109-145. Ponsford, K., & Kinsella, G. (1992). Attentiona l deficits following closed head injury. Journal of Clinical and Experimental Neuropsychology, 14 (5), 822-838. Posner, M.I., & Peterson, S.E. (1990). The attention system of the human brain. Annual Reviews of Neuroscience, 13, 25-42. Reitan, R.M., & Davison, L.A. (1974). Clinical neuropsychology: Current status and applications. Washington, D.C.: Winston & Sons. Robin, D.A., Max, J.E., Stierwalt, J.A.G., Guen zer, L.C., & Lindgren, S.D. (1999). Sustained attention in children and adolescents with traumatic brain injury. Aphasiology, 13(9-11), 701-708. Sosin, D.M., Sniezek, J.E., & Waxweiler, R.J. (1995). Trends in death associated with traumatic brain injury, 1979 through 1992: Success and failure. Journal of the American Medical Association, 273(22), 1778-1780. Stavro, G.M., Ettenhofer, M.L., & Nigg, J.T. (2007). Executive functions and adaptive functioning in young adult attention-deficit/hyperactivity disorder. Journal of the International Neuropsyc hological Society, 13, 324-334. Talland, G.A. (1965). Deranged memory. New York: Academic Press. Taylor, H.G., Schatsneider, C., & Rich, D. (1992). Sequelae of Haemophilus Influenzae meningitis: Implications for the study of brain disease and development. In M.G. Tramontana & S.R. Hooper (Eds). Advances in Child Neuropsychology: Volume 1. (pp.50-108). New York: Springer-Verlag. Teasdale, G., & Jennett, B. (1974). Assessment of coma and impaired cons ciousness: A practical scale. Lancet, ii, 81-84. 44


Tombaugh, T.N., Rees, L., Stormer, P., Harrison, A., & Smith, A. (2007). The effects of mild and severe traumatic brain injury on speed of information processing as measured by the computerized tests of information processing (CTIP). Archives of Clinical Neuropsychology, 22, 25-36. Van Heugten, C.M., Hendriksen, J., Rasquin, S., Dijcks, B., Jaeken, D., & Vles, J.H.S. (2006). Long-term neuropsychological performance in a cohort of children a nd adolescents after severe pediatric traumatic brain injury. Brain Injury, 20(9), 895-903. Wassenberg, R., Max, J.E., & Lindgren, S.D. (2004). Sustained attention in children and adolescents after traumatic brain injury: Relation to severity of injury, adaptive functioning, ADHD and social background. Brain Injury, 18 (8), 751-764. Weiler, M.D., Holmes Bernstein, J., Bellinger, D. C., & Waber, D.P. (2000). Processing speed in children with attention deficit/hypera ctivity disorder, inattentive type. Child Neuropsychology, 6 (3), 218-234. Wilde, E.A., Hunter, J.V., Newsome, M.R., Scei bel, R.S., Bigler, E.D., Johnson, J.L. et al. (2005). Frontal and temporal morphometric fi ndings on MRI in children after moderate to severe traumatic brain injury. Journal of Neurotrauma, 22 (3), 333-344. Wilens, T.E. (2004). Attention-de ficit/hyperactivity disorder and the substance use disorders: The nature of the relationship, who is at risk, and treatment issues. Primary Psychiatry, 11(7), 63-70. Wilson, J.M., & Marcotte, A.C. (1996). Psychosocial adjustment and educational outcome in adolescents with a childhood diagnosi s of attention deficit disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 35 (5), 579-587. Yeates, K.O. (2000). Closed-head injury. In K.O. Yeates, M.D. Ris, & H.G. Taylor (Eds.). Pediatric neuropsychology: Research, theory, and practice (pp.92-116). New York: Guilford Press. Zomeren, A.H. van, & Brouwer, W.H. (1994). Clinical neuropsychology of attention. New York: Oxford University Press. 45


BIOGRAPHICAL SKETCH Lindsey Boegehold received her Bachelor of Science degree in biopsychology and cognitive science from the University of Mich igan in 2005. She entered the Department of Clinical and Health Psychology at the University of Florida in 2006, taking a research assistant position in the pediatric neurops ychology laboratory of Dr. Shelley Heaton. She has also pursued research interests in th e area of womens health psycho logy under the mentorship of Dr. Deidre Pereira. After completion of the requiremen ts for her Master of Science degree in clinical psychology, she plans to enter the field of nurse-m idwifery. Eventually, Ms. Boegehold plans to combine her training in psychology and nursing to pursue research in the interdisciplinary field of behavioral perinatology.