Practice Guidelines forDirect Attention Training

McKay Moore Sohlberg, Ph.D.

Communication Disorders & S ciencesUniversity of Oregon

Eugene, Oregon

Jack Avery, M.A.

Department of Communication DisordersUnioereity of Minne sotaMinneapolis, Minnesota

Mary Kennedy, Ph.D.

Department of Communication DisordersUniversity ofMinnesotaMinneapolis, Minnesota

Mark Ylvisaker, Ph.D.

Department of Communication DisordersCollege of Saint Rose

Albany, New York

Carl Coelh o, Ph.D.

Communication Sciences DepartmentUnioereity of Connecticut

Storrs, Connecticut

Lyn Turkstra, Ph.D.

Department of Communication SciencesCase Wesf.em Reserve University

Cleveland, Ohio

Kathryn Yorkston, Ph.D.

Department of Rehabilita t ion Med icineUniversity ofWashington

Seattle, Washington

J<»,n..,1<>f M<dk<.l Spft<~ · I"v'(i'_ AuMl'¥l:l\'olu""" II. Number 3. PI> . i.-uxi:lCopyri, n' Cl 2003 by no,l",,,, Le~mi"ll. a d;v;. iun ofThon>tO<l !.earning. 10K.

xx ANCDS BULLETIN BOARDIVOL. 11, NO. 3

Th is a rticle is part of a llllncs of reports from a committee charged with dQVcloping evi­dence -based practice lEBP) guide lines fo r rehnbihteticn of cognitivc-cQlnmunicationdeficits following t rauma tic brain injury (TBn. We examine the literature for evidenceof the erteeuvecess of direct attention t raining to treat attention hnllainncntil followingTBI. Evidence is gleaned from the outcomes of nine Class I and Ctaee 11 studies thatspan interven tion from acu te to outpat ien t rehabilitation. Hcsultll and diecuaaion areor,g"a nizcd us ing five key quest ions aa a mechanism to rev iew the research to det ermineif the upprcach, outcomes, anti associated recommendations warrant II cha nge in clini­ca l practice. The key questions are: Who life the participan\.!l who received the inter­vention? Wha t comprises the attention trai ning? What are the outcomes of tilt" inter­veutien? Are there methodological concerns'! ~ there clinically <t pflhcable trendsIIC TOSS different attention remediat ion studies? The complexities nnd difficult ies inher­em in imp lem en ting clinical t rialli. wit h the heterogeneous TBI population a re dis­cussed . The a rt icle concludes with treatment guidelines and eptlcos suppo rted by theresearch review, Future eeaeareh needs arc highlig hted.

In recent years, the leadership of the Academy ofNeuro logic Comm unication Disorde rs and Sci­ences (A..NCDS), the American Speech-La nguage­Hearing Associa tion (ASHA), the Special InterestDivision 2 (SID 2-Neuro physiology and Neuro­genic Speech and Language Disorders ) of ASHA,and the Vete rans Administration recogni zed thetrend toward refe rencing research evidence to sup­port clinica l decision making in the ma nagementof medical conditions. In 1997, ANCUS emba rkedon the ta sk of establishing committees of expertsto develop evidence-based practice (EBP) gu ide­lines for the following areas; dysarthria, aphasia,de mentia. a praxia , and cognitive rehabilitation fort ra umatic brain injury. This article is part of a se­ries of reports from the commit tee charged with de­veloping EBP guidelines for rehabilitation of cogni­tive-communicative defici ts following traumaticbra in inj ury. For an overview of the committeeprocess, the reader is referred to the project intro­duction and initial com mittee report (Golper et al .,2001; Kennedy et al., 2002)

The in itial task of our subco mmittee was to ou t­line our philosophy and assumptions relevan t toEBP a nd clinical decision making, This required usto define "evidence." Evidence is the reason a clini­cian would pursue a specifi c treatment decision.Central to our posi tion is the proposition tha t thescience of clinical decis ion making for complex anddiverse populations includes, but is not restricted to,informat ion generated by randomized con trolledt rials and other clinical experiments. We describedour bas ic premises in an initial report and positionpaper (Ylvisa ker et aI., 2002). An early task was todefine the project scope by identifying the major ar­eas within the field of cognitive-communicative re-

habilita t ion wem for which we would developguidelines. These a reas include cognitive-communi­cation assessment, at tention training, managementof memory impa irments, social skillslbehavior regu­lation management, and in terven tion for metacog­nitive and executive function deficits. This currentarticle addresses attention training. Specifica lly, wediscuss the evidence relevant to t reatment ou t­comes associated with direct or s t ructured atten tiont raining fo r attention impairments followin g TBL

THE EVIDENCE·BASEDPRACTICE MOVEMENT

In an attem pt to establish defensible, vali datedintervention pra ct ices, the field of cognitive rehabil­itation has joined the healthcare movement's vigor­ous commitment to developing EBP. The primarygoal of this bu rgeoning reform is to ensure thai clin­ical decisions are guided by empirical evidence a nd,ideally, evidence from well-controlled s tudies thatsystema tically evaluate ou tcome, efficacy, and effec·t ivenees of specific interventions (Golper et at,2001). More specifica lly, the charge is to develop evi­dence-based guidelines that causally link treatmentprotocols to expected clinical outcomes (Robey,2001). Increasingly. however; it is recognized tha tstrict adherence to a preordained experimentalmodel as a filter to accept or reject particula r clini­ca l pract ices in a heterogeneous populat ion te.g., theTBI population ) may result in unsuccessfu l gener­alizations from resea rch to treatmen t outcomes. Sci­entifically sound clinical decision making includesa na lyzing exist ing experimental evidence in add i.­tiou: to a diverse set of individual patient coneidera-

PRACTICE GUIDE LINES FOR DIRECT ATIENTION TRAINING

lions and possibly the result of client-specific exper­imental intervention trials (Montgomery & Turkstra,2003; Ylvisaker et al., 2002).

The development of EBP guidelines for CCR ingeneral , and specifically for direct attention train­ing, required us to become conscious of our as­sumptions and biases relevant to analyzing reha­bilitation resea rch. This process made us mindfulof th ree fundamental be liefs. F irst, when reviewingthe literature, we realized the narrow question"Does it work?" oversimplifi ed the com plexit ies be­h ind direct at tent ion train ing delivered to TBr s ur­vivors. Second, we found it crit ical to remember thebidirectional nature of the relationship betweenpractice and resea rch . Third, we maintain that in­tervention outcomes are rela tive to persona l goalsand values.

Researchers are cha rged with the task of con­ducting clin ica l research tha t effect ively eva lua tesspecific aspects of particula r interventions in lightof previous wor k. To make well-founded clinicaljudgments, practitioners are challenged with therespons ibility of integrating existing research find­ings with their own clinica l experience, while re­specting the unique cha llenges presented by theirclients. We boldly suggest tha t it may not be possi­ble to answer t he question "Does a ttention remedi­a tion work?" using an unequivocal yes/no form at .Such a question assumes that attent ion remedia­tion is a uniformly delivered intervention that canbe compared across studies aud that the myriad ofrelevant contextual variables presen t in a clinicalpractice can be dismissed.

Our concern is tha t. the unequivocal "Does itwork?" question is divisive and encourages a battletha t is ne ither helpful nor illumi nating. Fordecades, the field of CCR seems to have beentrapped in all interna l debate over whether it is bet­te r to focus on training processes, skills, or complexfunctional abilities, and in what ways and in whatcontexts that training might best be accomplished.Although the battle is not over, it is no longer dis­puted tha t functional changes must be the goal oftreatment and that there are many ways to facili­tate functional changes. with an equal number ofways to measure them. The field continues to strug­gle with a quagmire of measurement issues. Wehave. however, learned that a one-size-fits -a ll solu­tion does not work. Individuals and fami lies re­spond to different interventions in different waysand at different times after injury. The response dif­ference is in part demonstrated by the widely diver­gent treatment results reported in the literature,which range from no change in performance aftertreatment to significant improvements (Sohlberg &

Mateer, 2001). As described in this current article,the wide swings of outcome in the attention reme­diation literature suggest tha t the Question "Does itwor k?" may be better replaced by "When does itwork best and for whom?" Then we can move on tosuch questions as "Is it worth the effort?"

No one disputes the va lue gained when researchadvances and im proves clinical practice. However,the investigation process is not unidir ectional.Clinical practice a lso sha pes research questions.The patient-specific hypothesis testing CPSHT) ap­proach discussed by Ylvisaker and Feeney ( 998 )reinforces the symbiot ic relationship between re­search and practice. Neither service delivery norapplied research exists in isolation. Clin icians relyon well-founded research to genera te clinical proce­dures that can be defended and supported withintheir delivery contexts. They must a lso often em­ploy diagnostic therapy to determine treatmen t e f­fects in specific pa tients. In turn , researchers' studyofcur rent clinical practices generates new researchquestions and highlights issues relevan t to treat­ment efficacy, particularly when the research repli­cates the rea lities of clin ical delivery (e.g., pa tientdemographics, dose, and duration limits). This iter­ative process is critical to the development of EBP.

Oue important assumption in our view of devel­oping EBP from existing Literature is tha t thisprocess necessarily draws on personal goals andvalues. Montgomery and Turkstra (2003) statedthat a problem with developing evidence-basedguidelines is that wha t constitutes a meaningfulclinical outcome is a personal and social judgmen t.As s uch, the ability to scru tinize the lite rature fo revidence will never be a wholly scien tific process inthe traditional sense of science. The ever-evolvingdefinitions describing different types of outcomespromoted by the World Health Organ ization(WHO) highlight the fact that this construct is a re­flection of social va lue. Consider the scenario of twoCCR experts evaluating a research outcome for anarticle investiga ting the effectiveness of a ttentionremediation. One analyst may examine the im­provements in neuropsychological test scores fol­lowing attention training as proof of treatment s uc­cess, while the other interprets it as a dismissabletreatment artifact . This is indeed the case withint he research (Par k, Proulx, & Towers, 1999 ;Sohlbe rg; McLaughlin , Pavese, Heidr ich, & Posner,2001). Acknowledgment of the subjective and val­ue-laden nature inherent in analyzing outcomes iscr itical to tile development of useful guidelines thathelp clinicians nav igate th rough the literat ure.

xxu

ESTABUSHING GUIDELINES FORATTENTION TRAINING

Establishing practice guideli nes for attent iont ra ining is a curren t foc us in the field of CCR (Ci­cerone et aI., 2000). The committee adopted a broaddefinition of a tten tion. We were interested in inter­ventions that addressed a wide assort ment ofskills, processes, and cognitive states tha t rela te tothe ability to focus and process incomin g informa­tion. At first blush, it may appear a st raightforwardprocess to measure and analyze the attention out­comes in individua ls with brain injury who have re­ceived attention training. Unfortunately, close in­spection revea ls thi s is not the case.

Direct Attention Training as an Interve ntion

Attention training is based on the premise that at­tent iona l a bilit ies ca n be im proved by act ivatingparticula r aspects of attention through a stimulusdrill approach . The repea ted stimulation of etten­t ional systems via graded a ttention exercises is hy­pothesized to facilitate changes in attentiona l func­tioning (Cicerone et al. , 2000; Sohlberg & Mateer,2001). Most a ttention training program s assumetha t aspects of cognition can be isola ted and dis­cretely ta rgeted with t raining exercises. The as ­pects of attention tha t are addressed vary widelyamong interventions and freq uently depend on at heoretical model of attention. Atten tion models,regardless of their opera tiona l framework, appearto include funct ions related to susta in ing attentionove r time (vigi la nce), capacity for inform ation ,s hifting attention, speed of processing, and screen­ing out distractions. Some a ttention efficacy stud­ies eva luate a ttention interven tions that focus onparticula r attention components s uch as reactiontime and sus ta ined a ttention for visual information(e.g., Pons ford & Kinsella , L988). Ot her efficacyst udies use a ttention training programs that in­cl ude hierarchical tasks to address a continuum ofattention components from basic sustained at ten­tion to more complex men tal control (e.g., Park etal . 1999; Sohlberg et al. , 200 1).

Tradi tiona lly, studies employ a un ique battery ofmeasures to assess possible changes related to in­tervention. Some studies limit their outcome mea­sures to individually selected neuropsychologicaltests te.g., Park et a l. , 1999; Sturm, WiJlmes, Or­gass, & Hartje, 1997). Alterna tively studies mayutilize patients' percep tion of change (e.g., Ci­cerone, 2002; Sohlberg et a l., 2(01) or their tas k

ANCDS BULLETIN BOARDNOL. 11. NO. 3

performance in an activity assumed to require a t­tentional processing (Kewruan et al., 1985) as anindicator of attention functioning. The disparateoutcome markers render it difficult to equate suc­cesses and therefore compare interven tions.

There a re severa l additional complications thatmust be considered when reviewing the attentionliterature. One conside ration is the rich opportuni .t ies provided by cross-populat ion inferences.As dis­cussed by Ylvisaker et a l. (2002 ), there is a wealthof information on populations who are neurologi­ca lly andlor symptomatically close to individualswith brain inju ry. Although our review is limited tothe a ttent ion train ing literature on individualswith traumatic brain injury, we wa nt to acknowl­edge extens ive informat ion relevant to attentionremediation available in other literature domainssuch as special ed ucation. Addit ionally, from anecological standpoint, it may be remiss to eva lua teone particula r approach for addressing ettentionalim pairments when in reality cl in icians often com­bine interventions. For example, direct tra ining isoften pai red with pharmacological management,training in the use of extern al aids, and/or meta­cognitive train ing.

Scrutin izin g the Atten tionLiterature Using Key Ques tions

As discussed , we submit tha t answering a binaryquestion querying whether attention training doesor does not work is not possible, give n the complex­ity of client, trea t ment, and research variables cou­pled with the value-laden nature of ou tcomes. Weoffer, as an al te rnative, a templa te of five key ques­tions tha t professionals can employ to eva luate theliterature. This approach offe rs a mechanism to fi l­ter the past , cur rent, an d fu ture a ttention inter­vention lite rature and help clinicians decide if theapproach, outcomes, and associated recommenda­tions a re a "good fi t" for their practice. The hope isthat the templa te will al low clinicians to scrutinizethe research to determine if research results war­rant a change in clinical practice and allow them todevelop an evolving portfo lio of ev idence-basedtreatment practices . We further hope that the tem­pla te will assist resea rchers in designing studiestha t reflect clinical reali ties. Key guiding questionsfor examining the attention training literature a re:

1. Who a re the participants who received theintervention?• What is the diagnosis or etiology?,. What is the injury severity?

PRACTICE GUIDELINES FOR DIRECT ATTENTION TRAINING xxnr

• What is the participant's age?• What. is the participant's level of education?• What is the time postonset of injury?• Are there dual diagnosis or comorbidity factors?• Wha t is the participant's cognitive profile

postinjury?2. What comprises the attention training?

• Wh at are the focus and rationale?• What are the treatment duration and frequency?• Where is the treatment setting?• Who are the providers?• Are training programs personalized to match

client skills andlor needs?• Are other in tervent ions incorpora ted into or in

addition to the delivery of attention exercises(e.g., reinforcement or strategy training)?

3. What are the outcomes of the intervention?• Are there measures suggesting changes in

at ten tion impairment (e.g., psychometric tests)following t rea tment?

• Are there measures suggesting changes inactivity/participation (e.g., changes in an atten­tion-demanding skill such as driving, changesin perceptions/rating of abili ty by client and/orcaregiver)?

• Are reported changes clinically meaningful?• Is there maintenance or generalization of any

reported changes?4. Are there methodological concerns? Are there

other explanations for given outcomes, and mayresults be either exaggerated or hidden?• What is the study design?• Are treatments compared to an alternative or

no treatment condition?• Are reliability and/or validity issues addressed?

5. Are there clinically applicable trends across dif­ferent a ttention remediation studies?• Are there ro bust findings tha t warrant a

change in practice?

A retrospect ive application of these key ques­tions to a former cognitive re habilitation interven­t ion outside of attention remediation serves as anillustration. Conside r the example of memory drillsthat were a comm on memory im pairment inter­vent-ion during the 1980s. Using this a pproach ,clinicians gave clients mult iple tria ls of list learn­ing or paragraph listening tasks as a means to im­prove recall a bility (e.g., Glisky & Schecter, 1986).Even tua lly, this a pproach fell out of favor whenboth clinicians and researchers found no improve­ment in either clients' or s ubjects' mem ory func­tioni ng. Had clinicians su bjected the literature tothe proposed quest ioning sequence in the template,

the field may have been q uicker to discard an inef­fective t rea tmen t. Specifically, if clinicians had fo­cused their reading of the litera ture on the mea ­surement outcomes (q uest ion 3), they would havenoted tha t memory dr ills occasionally resulted inimproved list learning for specifi c lis ts, but thatgeneralization was poor and no improvementswe re seen in other memory meas urements (God­frey & Kni ght, 1985; Moffat , 1992). Using the WHOoutcome nomenclature, there was :::I specific im­provement in a constrained activity, but no changesin the underlying impairment or activities/partici­patory performance on rela ted activities.

What Do E xist ing E viden ce-Base d Repor tsReveal About Attention Rem ediation ?

At the t ime of this art icle, there are two compre­hensive reviews examining the attention remedia­t ion literature (Cice rone et al., 2000; Park & Ingles,2001). Park and Ingles (200 1) reported on a meta­analysi s of the attention rehabilitation literature.They coded a nd analyzed 26 studies as direct re­training studies and concluded that subject perfor­mance im proved significantly on tasks tha t weretrained, but found no evidence of t rea tment effectson tasks that were different from those trained. Forsome of the st ud ies re porting a statist ical ly sigmfi ­cant im provemen t in one or more measures of at­tention (e.g., Gray, Robertson, Pentland , & Ander­son 1992; Niemann, Ruff. & Baser, 1990; Park et al.,1999; S turm et al., 1997), Pa rk and Ingles suggest.­ed that the pattern of im provemen t was a tt r ibu t­able to the acquisit ion of specific sk ills rather thanto the t r ain ing of attention . It is not poss ible toscrut ini ze their findings using our five key ques­t ions beca use the research methodology conscli­dates studies with different subject profiles, t reat­me nt a pproa ches, and outcome meas ures. Forexa mple, ma ny of the participan ts in the reviewedstudi es had severe bra in injuries result ing in im­pairment to basic at tentiona l processing; but be­ca use grouped data are presented, it is difficult toexamine individual profiles. Similarly, the groupeddata render it difficult to determine potent ially im­portant intervention characteris t ics. For exam ple,the t reatmen ts they a na lyzed were admin isteredan average of 3 1.2 hr; but with a standard deviationof 32.7 hr. Nonetheless, this s tudy reminds us of theimporta nce of analyzing the potentia l transfer ofa ny observed t raining effects to related tasks. It al­so encourages the field to a ddress definit iona l am­biguities such as the dis tinction between attentionskills and attention processes . Further, the meta-

m v

analysis highlights the question of effect size-ift here are observed changes in eit her skill orprocess performance, we must look at the magni­tude ofchange to determine clinical relevance.

Cicerone et a l. (2000) reported the findings of as ubcommittee of the American Congress of Reba­bilitation Medicine (ACRM) that ana lyzed existingresearch addressing CCR interventions for personswith TBI and/or st roke. They selected both t reat­ment effi cacy studies (highly constrained, time-lim­ited research with mostly homogenous samples)and studies of clinical effectiveness (empirical t reat­ment evaluations within clinical settings, which re­flected the actual use of an intervention). They useda screening process resulting in the select ion of 171articles for inclusion in their review. Thi rteen ofthese s tu dies were assigned to the ca tegory of re­med iation of a tten t ion deficits. Of these 13 st udies .3 were classified as well-des igned, prospect ive, ran­domized con trolled tria ls (Class I studies ); 4 wereclassified as us ing prospective, nonra ndcmized con­trols or a clinical series with contro ls (Class II stud­ies ); and 6 em ployed single-s ubject design or clini­cal re ports without controls (Class UI studies).

Trends across studies were sum marized, describ­ing the clinical implications of their literature re­view (Cicerone et a l., 2000):

Evidence . . . supports th(~ effecnvenese of attention train­ing beyond the effects of nonspecific cognitive stimulationfQr subjects ....-ith TB I or stroke during the poetecurc phaseof recovery lind rehabilitation .. . Interventions shouldInclude traini ng with different stimulus modali ties. levelsI)f complexity lind response demands. The interventionshould include therapist aeti"'iue8 such as mooitoring eub­jecte' performance, previdlng feedback, and teaching lIu at­egies. Attention traini ng appears to be more effective whendirected at improving the !lubject:lI performance on more

complex, fun ctional tusks. However, the effecte of t rent­ment may be relatively sm all or Ulsk specifi c. and an eddi­t ional need exists to examine the imllUet of attention treat­ment of ADL,I; or fun cnonnl outcomes. (p. 1600)

Summary of Stu d ie s I nclude din Existing R eview P a pers

In this sect ion, we utili ze our fi ve key questions asthe organizing matrix to review the fi ndings fromthose st udies reported to ha ve the s t ronges t. re­search methodology in the existing comprehensivereviews of at tention remedia t ion (i.e., the Class Iand Class fI st udies discussed in the reviews by Ci­ce rone et a l., 2000, a nd Park & Ingles, 2001 ). We se-

ANCDS BULLETIN BOARDN OL. 11, NO. 3

lected only those studies that specifica lly evaluatedthe d irect t ra ining ofa t tent ion to subject pools thatincluded survivors of traumatic brain inj ury. Table1 presents the Class I and Class Il st udies for sub­jects in the acute reha bilitat ion phase.Table 2 sum­marizes the Class 1 and Class II s tudies for sub­jects in the postacu te rehabili tation phase. Wereview this older body of literature collect ively a ndthen examine the recen t. litera tu re by d iscussingthe most recen t st udies individually.

Reliability for coding the six studies us ing thefi ve key questions was examined. The firs t t.wo au­thors independently reviewed four of t.he six stud ­ies (.66) and compared completed tables. There was100% reliability on key points , with two exceptions.There was not a consistent descript ion of the focusand rationale of treatment (key question 2) be­tween the two raters. Fur ther, for several s tudies ,there was disagreement abou t whether outcomes(e.g., responses to quest ionnaires) reflected impair­ment or activity/participa tion level changes. Withminima l discussion, consens us was reached onboth issues.

Who Recei ved the Interven tion ?

Cons isten t w it h ou r premise t.ha t read ers mus tscru t in ize resea rch to dete rmine its re levance totheir clinical practice, the wide va riety of s ubjectcha racteristics does not merge nea tly into a singu­lar pool of shar ed demographics. Two st udies (No­va ck , Ca ld well. Duke, Bergq uist, & Gage, 1996;Ponsford & Kinsella , 1988) focused on acute reha ­bili tat ion patien ts, and the rema ining four (Gray etal ., 1992; Niemann et a l., 1990; Sohlberg & Mateer,1987; Strache. 1987) focused on outpa tient clien ts.On ly three of the six a rticles (Nie mann et a l., 1990;Novack et al., 1996; Ponsford & Kinsella, 1988) re­ported on subjects with only TEl or closed head in­j ury (C J-U); the other st ud ies' s ubject pools includedunspecified "non-t ra umatic brain inj ury" (Gray etal., 1992), e VA (Strache, 1987), and a neurysm andpenetrating head injury (Sohlberg & Mateer, 1987).Three of the six studies exa mined performance ofsubjects wit h severe or ve ry severe inj u ries (No­vack et el ., 1996; Ponsford & Kinsella , 1988;Sohlbe rg & Mateer, 1987). On e s tud y's s ubjec ts(Niema nn et a l., 1990) spanned moderate to severeinj uries, and a nother study (Gray et al., 1992) de­scribed their participants a s having mild to moder­a te difficul t ies with a ttention, but d id not provideinformation on t he na ture or degree of actua l braininj ury. No inju ry severity information was providedin the sixth study (S trache, 1987).

PRACTICE GUIDELINES FOR DIRECT ATTENTION TRAINING

TABLE 1. e las!! I a nd Class n s tud ies for participants in t he acute re habilitat ion phase.

xxv

Reference

C LienblSu bje<:ta

Number of eubjecte

Cont rols

Diagnosis/et iology

Severi ty

Ago

Education

Time postonset

Dual dxlco-morbidity

Cognit ive profilepostinjury

Attention Trai n ing

Focus

Rauonale

Dura tion/frequency

Trea tment setting

Providers

Program ind ividualizat ion

Add itiona l or concu rrentcogni tive intervent ion

Ou tcomes

Impa irme ntJpsychom.change

Participation level changes

Methodology rev iew

Study design

Reliability

VaJidity

Novack e t aI., 1996ClaBs I

22 pairs

Random assignment to "focused" vs."uneu-uctu red" remediation

T81

"severe"

m '" 27 .8 (13.2) yrs VB. 26.4 (10.9) yn

m '" 11.5 (2.4) yra vs, 11.8 (l .G) yrs

m .. 5.9 (3.3) wks V8. 6.4 (4.9) wke

Not provided

Not provided

Focused vs. unstructured computerizedactivities. Focused tasks targetedvigilance

Focused program will promote moreextensive recovery of nttentional skills

m '" 20.5 (8.81 sessions va. 20.9 () 1.7)sessions . 30 min . 5 x wk

Rehab hospital

Masters level educator orpsycho metrician

Advancement for focused group basedon level of accuracy

Cuing d uring focused activit ies toensu re success

No s ignifica nt differences in attentionalskills or general cognitive abilities

No s ignifica nt difference of AOL orcognit ive FIMs

PTe- to posttrea t me nt compa rison

Not provided

Not provided

Ponsford & Kinsella, 1988Cla88 11

10

16 ortho pat ients (evalua tion nottreat ment controls )

CHI

""very seve re"

m '" 24.4 (8.7Jyrs vs. 25.8 (7.8) yra,

m '" 11.0 0 .9) yrs vs . 11 .8 (2.0) yrs.

range e 6-34 wks

Not provided

S ufficie nt senso ry a nd motor functionfor computer tasks

Com puter tasks: directed st imulationvs. independent work. Taskse mphasized vigilance

Cognitive benefit of boosting speed andselectivi ty of infor matioo processing

30 min for 15 days without feedback;30 min for 15 days with reinforce ment

Rehab hospital

Not provided

Feed back, reinforcement, and graphingof results for one phase

No s ignifi ca nt treatment effects; trendsnoted on single s ubject performancelevel

No s ignifica nt change on Rating scale.Ceiling effect seen on videomeas ureme nts

Mul tiple baseline across subjects

Not provided

Not provided

A general trend was seen for age, with the ma­jori ty of subjects falling between the range 0[ 25 to35 years. Two s t udies (N ie m ann e t al., 1990; S rr a -

m e, 1987), however, included subjects above theage of 60. Education trends were also seen. Thes u bjects of five s t ud ies had a t least 9 years of edu-

~. TABLE 2. Class I a nd Class II studies for partid panUi in the POSLacut.e rehabilitat ion phnse.

ReferenceGray e t a J. 1992Class I

Niemann e t al. 1990Class I

Soh lberg & I\laleer. 1987Class D

S t rache. 1987C lass II

Clien ts/Su bjects

Number of s ubjects

Cootrols

Diagnosislet iology

Severi ty

Age

Educat ion

Time postcnset

Dua l dia gnollislco-morbidity

Cognitive profilepostinjury

17 13 • Two groups of 15

I' 13 None 15

TB I or nOI1 ·'1'81 'I'Bt CHI, PHI, aneu rysm Head trauma , e VA, otheretiologies

"Mild- modera te to severe" Moderate to severe 24 hrs to i wecke LOC Not providedattenuonal dysfunction(not severi ty of injury)

m = 26.18 (7 .58) yrs vs . 28.9 (8.2) y ra VB. 25-.10 yrs. m = 32 yrs (range 20 to 70)34.14 08.44}yrs. 34 .3 (12.0) yrs.

Not provided 13.8 (l.8) yra va. 11- 13 yrs 9 or more yrs13.7 (:l.5) yrs.

Range.? weeks to 10 yrs 41.0 (2 J.5) months vs. 12- 72 months Less than 6 months to more37.1 (20.1) months. tha n 3 yrsRange : 12-72 mon ths

Not provided No eubatance abuse iss ues. Not provided Not providedNo premorbid psych.admissions

Subjective reports of poor DRS score of 100. VIQ a t 80-87; P IQ a t 74-98; Not providedconcent ration for real-life No severe a pahasia FSIQ a t 77-85situa tions

A t tention Train i.ng

F""",

Rationale

Microcomputer deliveredtask.: speed of a ttendingand information processing

Increased s kills will reflectimprovement in si milarneuropsych tests

Computer assisted attentionretrain ing program

Ccmputeriecd attentiontr aining will improvemeasures of atten tion butnot memory

APT and other commerciaJproduC1.S

Repeated st-imulation ofattention will improveimpa ired systems

Apparatus s upportedtraining ror deficits ofconce nt ra t ion

Increased attention skillswill improve physica l,psychological, social, andvocat ional process or reha b(ext ra polnted-c-not di rectlystated )

(con /i ll/US)

TABLE 2. continued.

Gray e t aJ. 1992 N iem ann et a l., 1990 Sohlberg & Ma teer, 1987 St rache, 1987Reference Clbs I Class I Class II Class U

Duration/frequency m _ 15.35 (2.06) hr over Two 2· hour sessions! 7- 9 training sessions! Twenty 30-min sessions3-9 wks VB. 12.7(3.8) hra wk x 9 weeks wk for 4-8 weeks ove r 4 wksrecreational comput ing

Trea tment setting Ou tpa tient/postacute Out patient program Post-acute day rrea i me r a Neuro logic rehabilitationclinic program center

Providers Not provided Not provided Not provided Trained assistants unde rinstruction ofneuropsyc hologist

Program individcalieation Not provided Advancement through Treatment tasks and degree Assigned to group re:menu of tas ks pe r of d ifficulty re: intake intake test ing. One grouppredetermined criteria testing. Advancemen t. per with s tanda rdized

individual criteria tra ining, one group wi thprogress-dependentad va ncement

St ra tegy t ra ining or Ye, Ye, Not provided Not providedfeedback

Outcomes

Impairmentlpsychometric improvement on 2 measures S ign ifican t improvement Sign ificant gains in Improvement for bot hChange of a ttention and in audito ry on 4 measures of a ttention. a ttention skills but not in groups beyond

.....orking memory at No treatment effect visua l processing s pontaneous recovery test6 months follow-up. generalization to repetition, or direct

dependent variables training of testedfunctions. ~Stight but d earadvanta ge" for thebaseline and progress-dependent grou p onmnemomic functions

Pa rticipat ion level changes Not provided Not provided All s ubjects improved in Not providedindependent living orreturn to work stat us

Methodology review

Study design Pre- a nd posttreatment Pre- and posttreatment Single subject mult iple Pre- lind postt reatmentcomparison ccmpariaon baseline across behaviors compa rison

design

~ .Relia bility Not provided Not provided Not provided Not provided

Validity Not provided Not provided Not provided Not provided_.

xxviii

cat ion, and two s tudies (Novack et e l., 1996 ;Sohlberg & Mateer, 1987) reported subjects with a tleast I year of college. Gray et al. ( 992) did not pro­vide premorbid education demographics.

Given the heterogeneous na tu re of survivors ofTBI. it. was st riking that only one of the s ix studiesprovided comorbidi ty information (Niemann et al..1990). Descriptions ofcogni tive profiles a t the onsetof trea tmen t varied widely. Specific markers suchas IQ (Soh lberg & Mateer, 1987) and subjective de­scriptions of distractibi lity (Gray et a l., 1992) wereused. Studies al so reported results of intake test­ing, but s pecific deta ils were often lacking and in­terpretation was left to the reader (e.g., Srrache,1987).

What Comp r ised the A ttenrian Remediation ?

Details of treatme nt specifics va ried widely amongthe six studies and ranged from specific descrip­tions of tasks (Pons ford & Kinsella , 1988) to gener­al trea tment profi les (Strache, 1987). Compute r­ized attention programs were used in a ll six studiesa nd were eithe r the sole focus of the t rea t ment(Gray et aI., 1992; Novack e t a l., 1996; Ponsford &Kinsella , 1988; Strache, 1987) or were supplement­ed wit h other tasks (Niemann et e l., 1990; Sohlberg& Mateer, 1987).

Studies also varied in the explicitness of descrip­t ions of the rationale and focus of t he treat ment. Inbroad strokes, each of the studies used exercises de­signed to stimula te discrete types ofattention, withthe expectation tha t improved perfor mance wouldfollow repetitive dri ll. The duration a nd frequencyof intervent ions, however, varied between the st ud­ies. Specific attention tasks in at least one studylasted for 5- 10 min (Nieman n et a l., 1990). Trea t­ment sessions ranged from 30 min (Novack et al.,1996; Ponsford & Kinsella , 1988) to 120 min persession (Niemann et a I., 1990) and varied from oneto two sessions da ily (Sohlberg & Mateer, 1987) totwice per week (Gray et al. , 1992; Niemann et nl..1990). Overa ll length of trea tmen t varied from 4(Niemann et al ., 1990) to 9 (Novack et al., 1996)weeks. As mentioned, trea tment occurred in acuterehabilitation settings for two studies (Novack etaI., 1996; Ponsford & Kinsella , 1988) and in outpa­tient clinics or settings for t he remaining four.Trea tment providers ranged from "tra ined assis­tants" (Niemann et a l., 1990) to teachers with grad­uate degrees (Novack et a l., 1996). No informationwas given abo ut the providers in the othe r fou rstudies.

ANCDS BULLETIN BOARDIVOL. 11. NO. 3

Individualized trea t ment plans fo llowing screen­ings or extensive eva luations were created in threest udies (Novack et 81.. 1996; Sohlberg & Mateer,1987; Strache, 1987). A hierarchy of exercises basedon assumed increases in difficulty was used in ansix st udies, although criteria for advancemen t tomore difficult tasks varied from ope ra tiona lly de­fined parameters (Niemann et nl., 1990; Sohlberg& Matee r, 1987) to individua l clinician decision(Novack et al., 1996) to ad vancement cont ingent onsuccess at each level of difficulty (St rache, 1987).Speed of processing was a key component in t hetasks of a t least four of the studies (Gray et el..] 992: Ponsford & Kinsella, 1988; Sohlberg & Ma­teer; 1987; Strache. 1987). 'Two st udies (Novack eta1. , 1996; Ponsford & Kinsella, 1988) focused on vig­ilance and selective attention ski lls, al though A

small percentage of Novack et al.'s subjects ad­vanced to alternating attention dr-ill. Subjects fromthe remaining four studies were exposed to alte r­nating and/or divided a ttention drill for at leastportions of their treatment protocol.

S tudies varied on reinfo rcement, feedback, ands tra tegy t raining. Novack et al. (1996) strategicallyad ded cues to guarantee subjec t success. Scores re­ported to s ubjects included formal reporting {Nie­mann et al.. 1990), discrete compute r screen dis­plays of task scores (Ponsford & Kin sella , 1988),comprehensive feedback (Niemann et al., 1990).and graphing of per formance changes over time(Ponsford & Kinsella, 1988). At least two studiesused tasks as a springboard for strategy training(Gray et nl., 1992; Niemann et al., 1990). One taskin Gray et al'e. study focused on overt "ver bal regu­la tion" train ing (i.e., coaching th e client to ve rballyself-cue as an attention strategy) during an a lter­na ting a ttention task.

What Are the Ouscomee of Intervention?

The s ix a rticles offer excellent opportunit ies to ex­amine the compl ica ted issues related to the subjec­tive va lues underlying the cl aim of a successful orunsuccessful outcome. In genera l, the studies tend­ed to employ a ba ttery of measurements to assesscha nges rela ted to interventio n. Before we ca nmake sense of the complexities behind outcome de­cis ions. we must discuss the differen t tools and a p­proaches used for measurement.

All six studies assessed some aspect of impa ir­ment via standardized testin g, and t his rangedfrom extens ive neuropsychologica l test batte ries(Gray et al., 1992; Sohlberg & Mateer, 1987; Stru­che, 1987) to a small number of s pecific tests sup-

PRACTICE GUmELINES FOR DIRECT ATTENTION TRAINING

plemented by speed and accu racy measurements ofnonspecified tasks (Novack et al ., 1996). Use ofthese cognitive measures also varied and rangedfrom pretreatment/postt rea tment. comparisons torepeat administration of specific tests over thespan of treatment (Ponsford & Kinsella . 1988;Sohlberg & Mateer, 1987J. Some studies used testsassu med to predict real world demands (Niemannet a l., 1990). Other tests were selected specificallybecause they were similar to treatment tasks (GrayeLal., 1992; Niemann et al., 1990). Four of the sixstudies used the neuropsychological tes t PASAT aspart of their measurement tools (Gray et al. , 1992;Niemann et al., 1990; Novack et a l., 1996; Sohlbcrg& Mateer, 1987), al though the manner of lest de­livery (e.g., taped presentation vs. live voice vs. re­vised tests) was not specified. Clear comparisonswith othe r outcomes are furt her tempered by thefact that. testing was incomplete (Novack et a t..1996) and that raw data were analyzed differentlybetween studies (e.g., Gray et a l., 1992; Sohlberg &Mateer, 1987). Other incons istent use of impair­ment leve l testing was also evident, Studies dif­fered in using parallel versions of repeated tests,including the use of versions supplied by the pub­lisher (Sohlberg & Mateer, 1987) or created by theresea rch administ rators (Ponsford & Kinsella,1988). One st udy created a unique way to score lestdata (ra t ios of accuracy scores to completion time)in an a ttempt to capture s ubjects' changes of per­formance that were not reflected in formal scoring(e.g., Novack et al., 1996).

Finding effective methods to measure relevantchange poses a sign ificant challenge. As discussed ,we need to understand the relevance of neuropsy­chological (i.e. , impairment level) testing for deter­mining success or fail ure of intervention. The re­port of positive gains following direct attentiontraining has been crit icized and dismissed as treat­ment artifact by some authors (Park et al., 1999),whereas others (Gray et al., 1992; Sohlberg & Ma­teer, 1987) specifically chose neuropsychologicalmeasurements fo r thei r simila rity to the trainingtasks and based their interpretation of findings ontest score improvement. Additionally, philosophicaldifferences are seen when results of cognitive testsare used to validate or reject the effectiveness ofthe training. Two studies (Gray et al., 1992; Nie­mann et al ., 1990) repo rted or implied poor gene r­a lization oftraining skills when secondary tests didnot improve over the pretreatment-posttreatmentcycle. Sohlberg and Mateer (l987) viewed the a b­sence of change in an untreated skill area (i.e.. the

double disassociation observed with unchanged vi ­seal-processing scores and improved attentionscores) as positive proof of the effectiveness of at­tention skills training.

In addition to the impairment level testing, threes tudies incor pora ted activity/participation level in­fo rmation in their assessments. The Functional In­depende nce Measure was used on a subset of onestudy (Novack et al., 1996), Ponsford and Kinsella(1988) created a rating sca le of dist ractibility andalso scored subject distractibility during an un­structured work task. Sohlberg and Mateer (1987)provided anecdotal info rmation about communityreintegration status after t reat ment was delive red .

Although it may be impossible to combine theoutcomes in to a binary decision on the efficacyand/or effectiveness of a ttention training, it is help­ful to look a t t he outcomes of the individual studies.The acute rehabiliuntion studies <Novack et a l.,1996; Ponsford & Kinse lla , 1988) reported improve­ment in their subjects, bu t based on control groupperformance or subject. changes during baselinephases, attributed it to spontaneous recovery. andnot to treatment. Both studies' participation leveltesting fai led to reflect benefit, although ceiling ef­fect problems were reported for Ponsford and Kin­sella's use of videouaped measurements.

Successful postint.ervention improvement wasreported in the ou tpatient studies. Gray et al .(J992) reported their subjects s howed improve­ment in storage an d manipulation of numericalmateria l in working memory, bu t. the s uccessemerged at follow-up testi ng only. They reportedimprovements in pictu re completion and speed ofprocessing skills, but admitted that cont ributionsof premorbid IQ and length of time s ince injurycould not be ruled out. They a lso reported that pe r­fo rmance decline in the cont rol group could haveartificially s kewed the improvements of the expe r­imental group. Strache ( 1987) reported tha t sub­jects demonstrated progress in concentration, psy­chomotor function. and intellectual and memoryfunction, and reported even stronger memory im­provement for a s ubset of subjects whose progres­s ion through thei r t rea tment hierarchy was contin­gent on their accu racy of performance, and nota rbit rarily decided. Niemann et al . (1990) reportedtheir subjects performed better on four measures ofa ttention in their evaluation battery, including Tri­a ls B and a cancellation Lest. Sohlberg and Mateer(1987) reported success ful gains by all subjects onthe PASAT and anecdotal reports of success in re­tuming to work or independent living.

Despite the successes of the four studies, im pair­ment level improvement does not. easily transla teinto clinica lly meaningful improvement. Even ifimpairment test ing should prove to be ecologica llyva lid in predicting improvements in independenceor community rein tegration , clinician judgmentand client/family input would be requir ed to deter­mine if the cost-benefit ratio was acceptable. Thisposthoc information is not provided in the studies.Similarly, the re was a lmost no attempt to measuremaintenance ofeffects over time in any of the stud­ies. Decisions about clinical relevance and furtherconfidence in the success of the intervention wouldbe easier if activity/participation-level assessmenttools had been used.

Are There Methodologicalor Cli n ica l Concern s?

We acknowledge that it is easier to "cr iticize thanto do" and that a ll studies, regardless of flaws, con­tribu te to the development of E BP. None theless,savvy consumers mus t. ident ify t.he strengths andweaknesses of the research as they conside r incor­porating new trea t ment ideas into t.heir clinicalpract ice.

We classified the six studies into three Class I(Gray et al., 1992; Niemann et a l., 1990; Novack etal., 1996) and three Class n (Pons ford & Kinsella ,1988; Sohlberg & Mateer, 1987; Strache, 1987)st udies. The Class I studies centered on pretreat­ment versus postt rea t ment measurement compar­isons; although controls were used, none of thethree had a "no-trea tment" group. Novack et al.(1996) compared their experi mental ' focused stim­ula tion group" with an "uns truct ured interventionprogram ." Niemann et a l. (1990) compared their ex­per imental attention training group wit h a memo­ry training group in a repeated measurement de­sign, whereas Gray et a l. (1 992) compared a groupwho received computerized training to a group thatused computers fo r recrea tion only. The Class nstudies included a multiple-baseline , across-sub­jects design (Ponsfo rd & Kinsella , 1988; Sohlberg &Mateer, 1987). Strache (1987) used two experimen­tal groups and a control group tha t received "nor­mal clinica l intervent ion." The two experimentalgroups received additiona l computerized trainingbut with different advancement criteria (arbitrnri­Iy deter mined vs. continge nt on accurate perfor­mance). Subjects were assigned to the groups basedon performance in "extensive diagnostic pretest­ing." In a ll the studies, the absence of a no-treat­ment group may undermine the r igor of the design,

ANe DS BULLETIN BOARDNOL. 11, NO.3

especially when Questions of spontaneous recoveryor trea tment effects are raised .

All six st udies referenced their attempts to con­trol for the many variab les that complicate clinicalresearch for TBI subjects. Typica l controls includedgender, age, severity, education, premorbid IQ, andso forth. Time postoneet was a lso routinely ad­dressed, and markers ranged from weeks tomonths to years. As previously noted, the studiesreported no improvements in attention training inthe acute rehabilitation phase. Inspection, howev­er, shows that one st udy (Ponsford & Kinsella ,1988) included acute care subjects who received in­tervention between 7-8.5 months following theirinjury. This time frame over laps with that of sub­jects of a t least two of the four postacute studies(Gray et al ., 1992; Strache, 1987).

Further methodological anomalies are noted inthe acute rehabilitation studies. Novack et al. (996)were unable to collect baseline measurements ontheir subjects with the exception of two tests, forwhich they devised "new methods of scoring" to cap­ture performance. Also, their participation levelmeasurement, FI}.f , was collected in the m.iddle ofthe intervention phase; it was unclear if the da tawere collected on 12 or 24 of the 44 subjects. Pons­ford and Kinsella (1988) created a ra ting s urvey ofattentional behaviors in da ily activities and incorpo­rated video-based analysis of distraction in a workmodule. Ceil ing effects undermined the effective­ness of the video task, and no informa tion is provid­ed on the design and reliability of the ra ting scale.Measurement issues also a rise when the absence ofnormative data is noted for the newly crea ted par­allel versions or one of their key measurements .

Finally, despite both studies' efforts at control­ling for differences between treatment and controlgroups, info rmation is not provided about the be­haviora l level or compliance of the s ubjects at thetime of the intervention, al though Pons ford andKinsella (1988) imply that their subjects were nolonger in posttra umatic amnesia . Even with con­trols for age. sever ity, and time postcnset, pa rti ci­pation in struct ured tasks could vary widely de­pending on t.he level of agitation, confusion, andfatigue. It is easy to imagine very different out­comes for s ubjects who received the same treat ­ment but differed greatly in their motivation andpotential for treatmen t participation.

Su bject recruitment and participation issuesmust. be addressed for the four outpa tient studiesas well. Although we are interested in attentiontraining delivered to surv ivors ofTBI, we must ac­cept tha t researchers in only one of the studies

PRACTICE GUIDE LINES FOR DIRECT ATTENTION TRAINING

used a subject pool comprised solely of persons withthat etiology (Niemann et al ., 1990). Subjects fo rthe other studies included survivors ofaneurysm orother CVAs (Gray et el., 1992; Sohlberg & Mateer;1987; Stracbe, 1987), those with penet rating hea dwounds (Sohlberg & Mateer, 1987), a nd those whohad othe r "neuros urgical procedures" (Gray et aI. ,] 992), Non-TBI survivors provided 50% or more ofthe subject pools of the th ree stud ies. Ifwe assumed ifferences in presentation of inj ury, pa rticipationin therapy, and courses of recovery between the eti­ologies, Doe must acknowledge that evidence out­comes resul ts may be skewed or a t best may be dif­fi cult to in terpre t. Controls' age may also need to beconside red. In Gray et al. (1992), the control groupis slightly older, but more important, the standa rddeviation spread is signifi cantly wider. Given theauthors' in terpretation that some positive resultsof the study may be caused by a declin e in the con­trol group's performa nce, this disparity must beacknowledged.

A disappointing methodological flaw present inall six reports reviewed in this subsection is thelack of inter- a nd in tra ra ter reliability. This infor­mation is lacking in the scoring of objective a ndsubject ive assessment tools, the collection of dataduring treatment sessions, a nd judgments rela tedto the interpreta tion of outcomes. Validity issuesare broader in scope and more diffi cult to quantify,ye t the a bsence of informa tion from the a uthors isonce agai n noted. To be valid clinica l intervent ions,the treatment tasks should be replicable across set­tings . The six studies here diffe r in the details pro­vided and range from a sentence-length descri ptionof the intervention (Novack et a l., 1996; Strache,1987 ) to a menu of treatment areas from which asubject's persona l treatment plan was created(Sohlberg & Mateer, 1987). Replicability of thespecifics of each of the studies would be impossible.Simila rly, re-creating the dose (e.g., 120-min ses­sions in Nie mann et al., 1990) or length of trea t­ment (e.g., 9 weeks in Gray et a l., 1992) fo r a clientwho was dependent on third-party payment of re­ha bilitation services might be di ffi cult.

Are There Trends Across S tud iesPubli.s h ed Prior to 1999?

Our review, in addition to Cicerone et al.s (2000) andPark and Ingles' (200 1), suggests there is evidence inthe literatu re of improvement in a ttention-basedskills with direct training; however, the studies thatreported improvements are open to interpretation.Exe rcises may promote the acquisition of specific

skills and outcomes may be task-specific (Ciceroneet al., 2000; Park & Ingles, 2001). The review of theolder a t tention literatu re and t he eq uivoca l out­comes directed us to scrutinize specific interventioncharacteristics and outcomes in more recent efficacyarticles. For instance, we looked to more contempo­rary studies to eval ua te whether treatment effectswere associ ated with. incorpora ting strategy trainingin the a t tention remedia tion, or whether those stud­ies that ma tched a ttention intervent ion tasks to in­dividual subject profiles had different outcomes t.hanthe studies tha t delivered a s ta nda rd program toevery subject. S imila r ly, we were in te rested inwhether more recent studies provided outcome re­ports that shed light 0 11 t.he task-specific nature ofattention training.

Recent Attention Efficacy Studies

A literatu re search was conducted to identify arti­cles published afte r 1999 (i .e. , after the Cicerone etal ., 2000 report) that eva lua ted the efficacy or ef­fectiveness of a ttention t raining. Data bases thatwe re sea rched incl uded PSYCH INF O, MED·LINElPubMed, Eric, and CfNAHL, us ing combina­tions of these key words: bra in injury, closed headinjury, a t tention, remedia tion, reha bilitation, andt ra ining. Twenty-seven a rticles were identi fied .These studies were reviewed a nd se lected based onthe following crite ria : (a) written in English, (b) ex­perimentally eval uated the direct tra ining of atten­t ion remedia tion to adul ts, (c) excluded s tud iesdealing with left. hemispa tial ina ttention, (d) s ub­jects included people with t raumatic bra in injury,a nd (e) outcome da ta were reported in the study.This screening process revea led three studies: Ci­cerone , 2002; Pa rk et a l., 1999; Sohlberg et al.,(2001) (see Table 3).

The rust two a uthors independently reviewedthe a rticles and coded them as Class II st udies. Thea rticles were reviewed using the five key questions.Again, there was high agreement between the tworeaders in the coding. The re was discussion abouthow to describe particular feat ures (e.g., whetheroutcomes were classified as impairment or partici­pation ), bu t the content of classification was consis­tent, and reliability was deemed acceptable. Tables1-3 summar ize literature according to the firstfour key questions. The fi fth key question (Aret here clinically a pplicable trends across the litera ­ture, is addressed in Table 4, with the generat ion ofpractice guidelines based on the evidence.)

Park, Proulx, and Towers (1999). These investi­gators evaluated the effectiveness of the commer-

TABLE 3. Recent s t udies lpost- l 999J.

Refe rence

Clie n tslSu bjeets

Number of s ubjects

Controls

Diagnoaisseric logy

Severi ty

Age

Education

Time postonset

Dual diagnosis!co-mo rbidi ty

Cognit ive pro filepostinjury

Attention Training

F'ocu.

Rationale

Duration/frequency

Treatment set t in g

Providers

Program individ ualiztion

St ra tegy trai ning orfeedback

Out com e s

ImpairmentJpsycbom.<hange

xxxii

Parka e t a l , 1999C la 8s 11

,.Culled fro m previouslycollected data. Selectivelyma tched for age andeducation

TBI

Sev ere

m = 37.3 yra (2.66 S E)

No repo rted d ifferencebetween groups

Less than Iyr to 4 ynJ

Not provided

"Slightly above average"

APT with hierarchi ca lexercises

Repeated s timula rien ofattention win improveimpaired systems

40 hours (media n) for7.2 mon ths (mea n ) "a bout2 hours" pe r session

Not provided

Psychologist

Repetition of aerciseswhen 3 or more errors.Adjunct counselin g re:salie nt cli nical issues

Yes.

No significant diffe rencein attention or workingmemory im provementsbetween 2 groupsNo change in depreeeicn .

Sohlberg e t aI., 2001Class 1J

1.

CTOS8Qver design

TBI, a noxia, tumor

LOC range : Null to 7months

m = 33. 1 yra va. 38. 1 yra

11 yrs ve. 13 yrs

l to 5 yrs

3 with cd iaeuee: 4 withmood medica dcna

Large range onneuropsychological testing

APT w ith h ierarchicalexercises

Repea ted stimulation ofa ttention will im proveimpaired systems

APT: 3 hrslwk X 10 wkB;education: 1 hTIwk X 10 wits

Un iversity clinic

Cert ified S LP orsupervised graduates tudent

Tasks chosen to m atcha ttention profile

No d uring APT ; yes d uringbrain injury ed ucation

Attention im proved efterAPT (PASA~ Stroop andTrails b for low vigilencesubjects). SignificantIm provement for aspectsof working memory

Cicerone, 2002Class U

••

TB I

Mild

m = 31 VB. 34 .75 yra

m = 15.25 yrs

m = 8.25 mon ths

Bxclueion for those withsignificant history

Im pa irmen ts on 216ad mi t teste pluss ubjective complain t

Working memory taskscombined with strategyt raining

Teaching conscious use ofstrategies to boostallocation of resourcesand manage speeddemands

60 mnlwk X 11- 27 wks

Outpatient clinic

Not provided

Timing and tasks variedret client need

Three qua rters of clie ntsimproved on attentiontests

(cont i n ues)

PRACTICE GUIDELINES FOR DIRECT ATTENTION TRAIN ING

TABLE S. (COll t i nucd)

xxxm

Refer en ce

Participation levelchanges

Meth odology review

St udy Des igo

~lial>i l it)l

Parks e t a l., 1999Cluss II

Not provided

Pre- postcompa riecn

Not provided

Not provided

Sohlberg et aI., 2001Cla ss II

Memory/attentionimprovements per surveysand in terviews

Crossover with in subjectexperimental design

Provided for rati ng scalesand interviews

Not provided

Cicerone, 2002Cla8s D

AJl of experimental groupresumed vocation orsocial roles

Pre- postcomparison

Not provided

Not provided

cially available Attention Training Program (APT)(Sohlberg & Ma teer, 1987). Da ta were presented 0 11

16 individuals with severe TEl who were beyondthe period of spontaneous recovery. Subjects wereidentified by "specia lists" as having a profile sug­gesti ng they would benefit from treatment. De­scri ptions of individual subjects were not provided .

Remediation consisted of struct ured API' exer­cises that were administered until a subject madeno more than two errors. Program administrationwas standard, and exercises were not selected tomatch a s ubject's ind ividual attention profi le.Treatment included the provision of feedbackabout performance and discussion of error pat­terns. Ai; the program proceeded, participants werealso educated about different types of at ten tion,and pa rall els between difficulties of daily livingand problems performing particular APT exerciseswere highlighted. Subjects typi caUy received twen­ty 120-min sessions spread over 29 weeks; thus , thetraining generally occurred less than once a weekfor more than 7 months.

Outcomes (question 3) were measured by com­paring pre- and posttreatment performance on twoneuropsychological tests (PASAT a nd ConsonantTrigrams). The PASAT was used because it is as-­sumed to be sensitive to attention impairments;however, the a uthors hypothesized that the recallmeasure of the Consonant Trigra ms would not beaffected by the types of attention addressed in theAPI' program. The Beck Depress ion Inventory wasused to assess the impact of attention training 011

mood. Results showed that the TBI group who re­ceived a t tent ion t raining improved on both of theneuropsychological tests, but not on t he Beck De­pression Inventory Measurement of maintenanceof effects was not addressed.

The study design (ques tion 4) compared pre- andposttraining test scores to test scores of age- andeducation-matched controls culled from data froman unrela ted research project undertaken 11 yearsprior to the study. These control s ubjects were ad­ministered the two neuropsychological tests on twosepa rate occasions over the course of a week in­stead of the 7-month in terval of the subjects. Thecontrol gro up data revealed improvement on thePASAT but not on the Consonant Trigrams. where­as the subjects with brain injury who had receivedAPT im proved on aspects of both tests. individualtest performance da ta were not provided; thus, it isnot possible to a nalyze performance within s ub­jects. The a uthors interpre t their fin dings to sug­gest that APT facilitated learning of specific ski lls,bu t not im provement of damaged a t tention func­tions.Reliability a nd va lidity concerns, and mea­surement issues, were not discussed .

S oh/berg, McLaughlin, Pavese, Heidrich, andPosner (2001). Th is group of investigato rs pub­lished an efficacy study of 14 postacute clients withmild-severe brain injuries. who exhibited impairedattention a bilities as determined by neuropsycho­logical evaluation. Group heterogeneity was report­ed, including histo ries of substance abuse, depres­sion, and/or ongoing lit igation.

ln tervention consisted of APT attention exercis ­es selected for each subject based on the results oftheir particular neuropsychological profile. For ex­a mple, a s ubject dis playin g particular di fficultywith s ustained and selective at ten tion worked ontherapy tasks designed to target these areas at aninitial level where he achieved 70-80% accuracy.Individu a ls received 24 hr of at tention train ing ad­ministered in three GO-min sessions each week fora total of 10 weeks. Attention drills were groupedby specific attention a bilities (e.g., s ustained etten-

XXXIV

tion ). Explicit performance st ra tegies were not pro­v ided , and instructions or activities to foster gener­alization to real world tasks were not offered. Sub­jects a lso received a single 60-min session eachweek devoted to brain injury education for thesame number of weeks as the APT atte ntion drillwork. Education consisted of a combination of top­ics selected from a menu of choices in combina tionwith supportive listening (a "check-in" for how theweek was going) a nd relaxa tion training.

Outcomes fo r the two intervent ion programs(APT and Brain Inj ury Educa tion) were measuredand compa red using both impairment- and activi­ty/participant-based measures. Impairment-levelmeasur es we re obtained us ing a battery of neu­ropsychological tests selected to assess different. a t­tention networks (vigilance, orienting, workingmemory, and executive functions ). Tests were ad­minis tered before and after each of the two inter­vention phases. Activity/participa tion measureswere obtained usin g sta ndardized questionnairesand stru ctured interviews to assess subjects' per­cept ions of t he ir neuropsychologica l and psychoso­cial performance in daily life.

A crossover design was used, in which hal f of thesubjects received the attention training prior to theBrain Injury Education and half of the subjects re­ceived the opposite orde r of treatment. This designa llowed subjects to be used as their own controls.Based on responses to structured interview andperformance on neuropsychological testing, im­provement in complex attention abil it ies te.g..working memory, alternating attention) were seenfollowing APT. In contrast, there was little s pecificimprovement in basic a ttent ion (e .g., vigilance ororienting abilit ies ) following APT administration.Brain Injury Education was most effect ive in im­provin g self-reports of psychosocial function . lm­proved PASAT scores were found to be correla tedwith self-reports of im proved attention on stru c­tured interviews. It was noted tha t vigilance levelinfluenced the improvement resulting from atten­tion training, in that those clients with higher ini­tial vigilance improved more on measures of execu­tive a ttention. The au thors in terpreted t heirfindings as suggestive that APT was effective inimproving working memory or complex attentionin clients with intact vigilance. They do not addressother possibilit ies fo r improvements, such as sub­jects learning some type of behavior or skill fromperforming APT exercises that resulted in in­creased performance on specific neuropsychologicaltests.

ANCDS BULLETIN BOARDNOL. I I, NO. 3

Reliability and validity issues regarding thestandardized questionnaires and structured inter­views were indirectly and directly addressed. Thethree surveys administered to the s ubjects werefrom previously published articles and, in two ofthe three, answers were collected from the subjectsand a member of their family. Specific informationon the structure of the interviews, the transcrip­tion, coding by najve readers, and data analysis a redesc ribed .

Cicerone (2002). This investigator recently evalu­ated t he effectiveness of an intervention designedto address attention deficits following mild trau­matic brain injury (M'rBI). Treatment pa rt icipantsconsisted of a convenience sam ple of patients re­ferred to a postacute brain injury rehabilitat ionprogram based on a diagnosis of MTBI. Four sub­jects rece ived a ttention remedia tion and fourserved as a control group based on thei r inability toreceive trea tment . Neithe r criterion for MTBI norindividua l s ubject data was presented. The subjectswere matched closely for age, gende r, educa tion,and months postinjury, with all a t least 3 monthspostinjury. AU s ubjects had to meet cri teria for sig­nificant impairment on two out of six attentionmeasures; however, the treatment group partici­pants were init ially tess impai red than the controlgrou p.

The treatment focus in this study varied fromthe other two intervention studies reviewed in thissection. S imilar to the Sohlberg et. at. (2001) study,the author employed hierarchically organized a t­tention remediation tasks targeting complex a tten­tion skills via working memory tasks that were tai­lored to match the specific attentioual profi les ofthe individua l clients. However, unlike both of theaforemen tioned studies, the focus was on using theattentional tasks as a method for training in theuse of metacognitive strategies such as verbal me­diation. rehearsal , anticipat ion of task demands,self-pacing, and self-monitoring. The interven tionemphasized the conscious and deliberate use ofsuch strategies to increase the participants' abilityto allocate their a ttention resources and control thepacing of task performance. Hence, although therewas repetitive administration of attention exercis­es, stra tegy training was a primary emphasis. Theschedule of trea tment was 1 hr per week for 11-27weeks.

Treatment outcomes included neu ropsychologicalmeasures, self- rating for perception of change, andinformal reports of changes in status fo r vocationaland social roles. Impairment-based measures com­pared pre- and posttreat ment scores on a number of

PRACTICE GUIDELINES FOR DfRECT ATIENTION TRAINING xxxv

attention tests with the performance of the DO-­

treatment control group. Res ults showed tha t threeof the four participants improved significantly onthe a ttention tests, but none in the control group didso. The trea tment group also improved significantlyon their se lf-re port of a greater reduction of atten­tion difficulties in comparison to no change in thecontrol group. The author anecdotally reported thata ll of the treatment group participants returned toprevious vocational a nd social roles, but none of thecomparison group participants did so during thesame period. Change was attributed to im provedst rategy use rather than im proved underlyin g at­tentiona! processing, al though there is no attemptto parse out these factors.

The study used a prospective, case-comparison de­sign with groups of fo ur individuals. StatisticaJanalyses were thus performed on a very small sam­ple. The author reported that the rating scale wasdeveloped for the study and that formaJ psychomet­ric review had not been completed before use. Relia­bility of measurements was not discussed. Individ­ual subject profiles were not provided; thus, it isdifficult to analyze poss ible threats to internal valid­ity. One threat to interna l validity resulted from thefact that the treatment. group performed better ini­tially on attention measures. which could indicatethey had more cognitive resources and, therefore.j ust needed the strategies to facilitate improvement.

S u mmary of R ecent. Effica cy S t ud ies

S ubject variabi lity was quite large across studies.Park et a l. (1999) stated that s ubjects had seve rebra in injuries, whereas the s ubjects in Cicerone'sstudy (2002) were all d iagnosed with MTBT. Thesubjects in the Sohlberg et 31. (2000) study spa nnedthe range from mildly to severely impaired. All sub­jecta we re reportedly beyond the period of sponta­neous recovery. All three studies ad ministered hi ­erarchically organized a t tent ion drills, with two ofthem (Cicerone, 2002; Park et al .. 1999) adding as trategy feedback component. One of the studiesreported a low intensive t herapy regime (less t hanonce per week), with a protracted time of service(Park et al., 1999). In the other two studies, sub­jects were treated at least weekly, with the subjectsin the Sohlberg et a l. (200 1) study receiving thera ­py 3 hr per week, whereas Cicerone (2002) treatedsubjects about 1 br per week. These same two st.ud­ies individualized the selection of a ttention tasks.

To measure outco mes, Cice rone (2002) a ndSohlberg et al. (2001) used im pairment- and activ­ity-based measures, whereas Park et a1. (1999)

used impairment measures only. All three studiesreported changes on attention tests in the group re­ceiving attention training. The interpretation ofthese findings, however, differs widely between thethree stud ies. Park et al. suggested that thechanges demonstrated by their subjects were notsignifi cantly di fferent from those in the nonbrain­injured control group and that the changes weremost likely d ue to specific practice on tasks that re­semble the outcome measures. Sohlberg et al . sug­gested tha t the profile of change on tests in theneuropsychologica l battery supports improved cog­ni t ive funct ion ing, specifically in complex atten­tion/executive fu nct.ions a nd working memoryprocesses. Cicerone cla imed tha t. the benefits of h istrea tment were due to pa rticipants' improved a bil­ity to compensate for residual deficits a nd adoptstra tegies for more effect ive allocation of their re­maining a ttentional resou rces .

The two studies reporting more robust changesfollowing trea tment, including improvement in sub­jects' daily functioning (Cicerone. 2002; Sohlberg etal ., 2000, shared the following features: (a) individu­alized attention exercises. (b) treatment sessionsthat were 1 hr (vs. 2 hr) in duration, (c) at leastweekly treatment sessions, (d) outcome measuresthat included a range of different tests sensitive toattention and working memory. and (e) outcomemeasures that included activity-based measures us­ing client self-report. data. Additionally. examinationof older literature in conjunct ion with this currentlite rature suggests that the inclusion of strategy ormetacognitive tra ining, as part of direct a ttentiontraining, increases trea tment effectiveness.

WHAT HAVE WE LEARNEDFROM THE LITERATURE?

The answer to th is q uestion lies in our in terpreta ­tion of the evidence. Unfortunately, studies evalu ­ating outcomes following cognitive intervention arenot as straightforward as pharmacologic st ud ies.The heterogeneity inherent in the TEl population,coupled with the s trengths and limitations uniqueto each se tting and practitioner and the range ofopin ions regarding what constitu tes meaningfulcha nge, makes it di fficult, perha ps impossible, todesign studies with clean, unequivocal outcomes. Itis hoped that because this a rticle has been writtenby a committee of resea rchers and clinicians incor ­porating a broad range of perspectives, the biastha t occurs when in terpret ing evidence will be tem­pered . In general , we fool confiden t in our assess-

xxxvr

ment that cer tain aspects of a t ten tion training a rehe lpful in improvin g attention performance insome adu lts with TBI.

Recogni zing the difficulty inherent in interpret­ing the evidence, our ta sk bas been to examine t hea tte ntion t raini ng literatu re for evidence that reosponds to the ques tion "when does attention train­ing facilitate the greatest cha nge and fo r whom?"We conclude with a set of recommendations fo r im­plementing direct atten tion train ing. Practiceguidelines are recommendations for patient man ­agement reflecting moderate clin ica l certainty, usu­ally evidence from Class H experiments or a strongconse nsus of Class III evidence (Miller, Rosenberg,Gelinas. & the ALS Pract ice Parameters TaskForce, 1999). The suggestions offered in Ta ble 4were generated from our review of the attentio ntraining literature. Given the uneven and incom­plete natur e of the experi mental literatu re , indi­vidua l studies addressed different clinical pract icequest ions. We selected aspects of our fi ve key ques­tions that we believe are sufficien tly s upported byevidence from one or more Class n stud ies to en­courage clinicians to adopt a particular clinica lpractice. Table 4 ehould be viewed as our interpre­tation of the current lite rat ur e in conjunct ion withour own collective experience as cognitive rehabili­tation practitioners.

Examining the literature for evidence that re­vea ls the source of observed changes in a tten tionperformance highligh ts the in terpre ta t ion chel ­lcnge. For exam ple, Park et al . (1999) hypothesizedthat the im proved performance on the ConsonantTrigrams test suggested changes in attention be­haviors or s kills rather than improved cognitiveprocessing. An alterna tive explanation may be thatth is recall measure requi res working memory,which had improved as a result of at tention tra in­ing. Converse ly, Schlberg et al . (200 l) claimed theirtraining resulted in improved processing; however.it is p lausible that the training may have in partencouraged adoption of a n assertive a tt itude or areduction in a nxiety while performing t he morecomplex tests and daily living tasks (Fasot t i, Ko­vacs, Eling, & Brouwer, 2000). In short, the ques­tion of why some indi viduals' performance changeswith the a tten tion training remains unanswered .

FUTURE RESEARCH DIRECTIONS

Our review process exposed and educated our com­mittee on the cha llenges to writing EBP as well as

ANCDS BULLETIN BOARDIVOL. 11. NO. 3

illuminated future research directions. Ooe of thedifficulties we encoun tered was how to code the rel ­evant fea tures of the effi cacy research. We fonnu­lated our committee with tbe notion of assemblingexpert reviewers who both provide cognit ive reha ­bilitation services a nd who cond uct research in thefield . Another option would have been to use blindreviewe rs to code the studies in order to achievemore objectivity in the review process. Instead , weelected to use d iscussion and consensus to identifya nd code relevant research features. Further, biaswas poss ible by having reviewers who have con­t r ibuted to the a ttent ion resea rch literature andwho have intellectual owne rship of the interven­tion under question. However, the committee a p­proach to this project provided checks and ba lancesto gua rd agai nst this type of bias. The re a re prosand cons of relying on "front-line" experience ver­sus objective a ppraisal.

A further limitation of our review comes froms umma rizing rather than providing the details ofsome of the important methodologica l research is ­sues. For exa m ple, we chose Dot to discuss in detailthe specific types of reliability and va lidi ty thatwere lacking in stu d ies . This decision was madedue to space considerations and a desire to disc ussa broad range of research issues.

We look forward to conti nued resea rch tha t. facil­itates efforts to develop evidence- based guidelinesfor attention train ing. The most outstanding needrevealed by our work is to develop methods tha tmeas ure the impact of at ten tio n im pa irments ondaily life for individuals across the lifespan. The re­la t.ionship between neuropsychological tests andattention as deployed during real world activiti es isnot clear. The increased availability of functionalbrain imaging may help identify neuropsychclogi­ca l circuits that a re a ffected by train ing, but again,we need lo understand how these circuits relate tothe perform ance of funct ional activities. It, may befruitful to study alternative assessment paradigmsused in related fields such as the functional assess­ment model descr ibed in special ed ucation researchand practice (e.g., Lucyshtn . Albin, & Nixon, 1997)or inter pretive research methods ada pted to eva lu ­a te the impact of impairments in an ind ividual'sdaily li fe (e.g., Simmons-Mackie & Da mico, 1996).

Our review process highlights the need for re­search that better describes the specific elements ofattention training that are most effective in partic­ular contexts and the outcomes that result fromsuch training. For exam ple, the knowledge gainedfrom our review encourages fut ure studies tha t in-

TABLE 4. Clinical recommendations based on t.he review of literature, organized by key questions 1 through 3.

PracticeKe y Ques t ion Recommendations Summary of Evidence Clinic ians Should

I . Who is a good Gu ideline for pcstacute Two CIa!iS I and four Scrutinize candidacy andca di d ate fo r direct or mildly injured c1ient.a. CIR8S " studies with monitor responses toa tte n t ion training? with intact vigilance descriptions of training

participants

Ins ufficie nt evidence to One Class I and one Scrutin ize cand idacy.make recommendation Cteee II s tudies. but with Monito r responses tofor clients at acute phase ques tionable internal t raini.ng a nd know thatof recovery va lidi ty (incomplete da ta , observed improvements

unre ported va ria ble may in part be a res ult ofcont rols. etc.I s ponta neous recove ry

Unknown for use with Evide nce provided by Be ca utious and awa re ofclients with severe ly incomplete acute care uncertainties of outcome.impai red vigilance studies (one Class I and Proceed on case-by-case

one Class II). basis

2. What are the c r it ica l Guide line for using di rect Two Class I and one Class Use attention training infeatures of direct attention t raining in " studies combination with self-attention training? conjunction with reflective logs,

metacognirive training a nticipat ion/prediction(feedback. self- monitoring, activities, feedback. && s trategy t ra in ing) st ra tegy training

Guideline for program Th ree Class 11 studies Identify client strengthsindividua liza tion and needs prior to

t reame nt; select exercisesto address specific areasof weakness

Guide lin e for t reat men t 1'wo Class I and four Administer treatment atfrequency Class II studies least once per week

Guideline for complex Four Class II studies Use a hiera rchy of tasksatten tion tasks that emphasize working

memory, menta l control.a nd select ive,a lternat ing, an d/ordivi ded attent ion

Unknown for im proving One Class I and one Class Be cautious when usingvigilance or reaction time II s tud ies wit h acute remediation programs

pa rticipants as described that focus on simpleabove vigilance or reaction time

S. What outcomes can Gu ideline for obtaining Two Class I and four Iden ti fy desiredyou expect. from task specific, impairment Class II studies outcomes. Measuredirect a tte n tion level ou tcomes perfonnancetraining?

Unknown for obtaining One Class 1 and two Identify desiredgeneralization to untrained. Class II studies but outcomes. Measureimpairment level tasks striking differences in perfonnance

results interpretation

Uncertain for obtaining Cri terion-refe renced Ide nti fy desired outcomes.gene ralization to ou tcomes in three Cteas Use mel-hods that canpa rt icipan t level tasks II studies reliably measure clin ically

meaningful progresa

xxxvii

xxxv...

vestigate the effects of combining st rategy-basedt rain ing and d irect attent ion training. More coo­trolled studies with larger numbers of participantgroups that are carefully described are also cri t icalThe use of control groups receiving serial measure­ments without any treatment would strengthen ou runderstanding of intervent ion effects, as would theuse of examiners who are blind to s t udy des igns a ndpart icipan t group assignment. Controlled studieswould help sort out questions s uch as the effective­ness of attention training for acute patients.

We recogn ize the challenge of fo rmulating treat­ment groups within the brain injury populat ion.Looking to populations that share similar feat ureswith the bra in injury population may mitiga te thisresearch cha llenge. For example, a recent studyeva luated the effects of a ttention training in a pedi­atric popula tion with attention deficits as a result ofbrain radiation for cancer treatment (Butler &Copeland, 2002).Their results were consis tent witha number of studies reviewed for this a rticle andshowed significant changes on neuropsychologica ltests measuring attent ion following attention train­ing, but the changes did not gene ra lize to a func­tional acade mic task. Widening our search to in­d ude other populations such as individuals withdevelopmental at tention deficits may be product ive.

It is our hope that summarizing future resea rchusing the five key questions presen ted in this a rti­cle may provide an organizational framewor k tha twill a llow clinicians to make decisions that a re em­pirically su pported.

Acknowledgments: The a ut hors wan t to thank allthe expert reviewers who provided valuable feed backon an earlier version of this manuscript. This projectwas sponsored by the Academy of Ne urologic Commu­nicatiou Disorders & Sciences fANCDS ), the Amer ica nSpeech-language-Heari ng Association (AS HAl, Ne uro­physiology and Neuroge nic Speech a nd Language Dis­orders (AS HA's Special In terest Division 2), and theVeteran Affairs Administration. Cor respondence con­cern ing this article s hould be addressed to Ma ry R.T.Kennedy, Ph.D., Chair, ANCDS Wri ting Committee, evi­dence-ba sed pra ct ice guidelines for cognit ive-commu ni­ca tion disorders after trau matic brain inju ry, Depa rt­ment of Comm un ica tion Disorders, University of Min­neso ta , 115 Shevlin Ha ll, 164 Pills bu ry Dr. S.E. , Mi n­nea polis, Minnesota , 55455.

Addre ss co rre s pond e nce to Mc.Kay Moore Sohlberg.Ph .D., Communication Disorders & Sciences, 528 ] Uni ­versity of Oregon, E uge ne, OR 97403, USA.e-ma il: mckay@oregou.uoregon .ed u

ANCDS BULLETIN BOARDNOL. n , NO. 3

REFERENCES

But ler, It W., & Copeland . D. It (2002). Attentionalprocesses and t hei r remed ia tion in ch ild ren t rea tedfor cancer: A literature review and the developmen tof a t herapeutic approach. Jou rnal of l nterna tionaiN europsychological Society. 8, 115--124.

Cicerone, K. D. (2002) . Remedia t ion of 'working a tten­tion' in mild tra umat ic brain inju ry. Brain Inj ury, 16,185- 195.

Cicerone, K D., Dahlberg, C., Kama r, K , La nge nbahn,D. M., Malee, J . F.. Bergquist, 1'. F. et a l. (200m . Evi­deuce-based cognitive rehabilitation: Recommenda ­tions fo r clinical pract ice. Archives of Physical Medi ­cine & Rehabilitation, 81 , 3 16-321.

Fasotu, L., Kovacs , F., Eling. P. A. 1'. M., & Brouwer, W.H. (2000). Time pressure management as a ccmpen­satory st ra tegy t rain ing after closed head injury.Neuropsychological Reha bilita tion, 10,47-65.

Glisky, E. L., & Schacter, D. L. (l 9861. Remediation oforganic memory d isorde rs: Cu r re n t st a t us a ndfuture prospects. -km rnol of Head Trauma Rehobui­lotion, 1, 54~63.

Godfrey, H. P. D., & Kn ight , R. G. t 1985). Cognitive reha­bi lita t ion a nd me mory functioning in a mnesi ac a lco­holics. Journal of Consulting & Clinical Psychology,53. 555-557.

Col per, L. C., We rtz, R. T. , Frattali, C. M., Yorkston , KM.. Myers, P , Katz, R., et al . (200 1). Eindence.bceedpractice gu idelines for I he management of comm uni­cation disorders in ncurologically impaired individu­als: Proj ect introduct ion . [On-line]. Availa ble: http ://www.ancds.duq.edulPracticeGuidelines

Gray, J . M., Robertson. I., PenUand, B., & Ande rson, S.( 992). Microcomputer-based a ttent iona l retra in ingafter bra in damage: A randomized group controlledtrial . Nenropeychalogicei Re tra ining, 2, 97-115 .

Kennedy, M. R. T., Avery, ,I., Coelbo, J. , Sohlberg. M. M.,Turkstra, L., Ylvis aker, M., et a l. (2002). Evidence­based practice guidelines for cognitive-communica­tion disorders a fte r traumati c brain injury: Initia lcommittee report. Journal of Medica' Speech-Len­HI/age Pathology, 10(2). ix-xiii .

Kewman, D. G., Se igerma n, C., Kinter, H. , Chu. S., Hen­son, D., & Redder, C. (]985). Simulation training ofpsychomotor skills: Teaching the brain inj ured tod rive. Reha bilitation Psychology, 30. 11-27.

Lucysh in, J . M., Albin, R. W.• & Nixon, C. D. (1997).Embedding comp rehensive behavioral support infamily ecology: An experimental, single-case a naly­sis. Jou rnal of Consulting and Clirr icfl l Psychology,65.241- 25 1.

Miller, R. G., Rosenberg, J. A. , Gelinas, D. F.. & the ALSPractice Parameters Task Force . (1999). Neurology,52, ] 311- 1323.

Moffat , N. (1992). Strategies of me mo ry the rapy. In R A.Wilson & N. Moffa t (Eds.j, Clinical ma nagement of

PRAC1'ICE GUIDELINES FOR DJREC1' ATTENTION TRAIN ING XXXIX

memory problems (pp. 83-116). Sa n Diego: SingularPublishin g Group.

Montgomery, E. 8. , & Turkan-a, L. S. (2003). Evidence­based medicine: Let's be reasonable. Journal of Med­ica l S peech-La ngunge Pathology, 11(2), i x-xi i .

Niemann, H., Ruff, R. M. , & Base r, C. A. (]990). Cern ­puier-assiated attention retrai ning in head-injuredindividuals: A controlled efficacy study of an outpa­tient program . Journal of COl/suiting lind ClinicalPsychology, 58, 8 11-817.

Novack. T. A., Caldwe ll, S. G., Duke, L.. Be rgqu ist , 'f. F.,& Gage , R J. (1 996"1. Focused ve rsus unstructu redintervention for a ttention deficits after traumaticbrain injury. Journal ofHeed '1Tuuma Rehabilitat ion,11, 52-60.

Park. N. W.. & Ingles ,J. L. (200 1). Effectiven ess of auen­tion rehabilita ti on after acqui red bra in injury: Ameta-a nalysi s. Neuropsychology, 15, 199-210.

Park , N. w., Proulx, G., & Towe rs, W. M. (1999). Evalua ­tion of t he {m e nt ion process training programme.Neuropsychological Rehabilitat i<1Jl , 9, 135-154.

Ponsford , J. L., & Kinse lla, G. (988). Evaluation of aremedial programmed for a ttent iooa l defi cits follow­ing closed head injury. JOllrrll1l QfClinical and Esperi­mental NeuropSJdwlogy, 10, 693-708.

Robey, R. (200 IJ. CEU part II I: Evidence-based practice.ASHA Special Interest Divi$ioll 2 NewslelJer, 11{]),10-1 5.

S lmmens-Maekle, N. M., & Damico. J . S. (1996).Accounting for handica ps in a phasia. Disability andRehabil itation, 18, 540-549.

Sohlberg, M. M., & Mateer, C. A. (1987J. Effectiveness ofan attention-training progrnm . Journal of Clinicaland Experimental N europsychology, 9,117-130.

Sohlberg, M . M ., & Ma teer, C. A. (2001). Cognitive rehn­bititation: All in teg ra ted neuropsychological ap­proach. New York: Guilford Pub licat ions.

Soh lberg, M. M., McLaugh lin, K. A., Pavese, A., Hei ­drich, A., & Posner, M. (2oo1J. Evaluation of etten­tion p rocess training and brain injury ed ucation inpersons with acquired brain injury. Journa l of Clinl­cal a nd Esperimetuai Neurop$)'chology, 22, 656-676.

Srraehe. W. (1987). Effectiveness of two modes of t rain ­ing to overcome defi cits of concen tration . Intem a­tianal Journa l of Rehabilitation Research. 10 <S up-­plemen t 5), 1418--1458.

Stu rm, W., WiUmes. K., Orgl\88, B., & Hartje, W. (1997).Do specific attention deficits need specifi c training?Neuropsychological Rehabilitation, 7, 81- 103.

Ylvisaker, lV1. , Coelho. C., Kennedy, M., Sohl berg. M. M..T urkan-a. L.. Avery, J., & Yc rkston. K. M. (2002).Reflect ions on evidence-ba sed pra ctice and rat ionalclinica l decision making. Journal ofMedical Speech­La nguage Pathology, 10(3), xxv- coodu.

Ylvisakee, M., & Feeney, T. (1998). Couabora tiue braininju ry interuenticn: Positive everyday routines. SanDiego: S ingu lar Publishing Grou p.