Original Research

Tr ryPRRM , PhKa

ObaftanDetatSetraPawetheM6 hMmeapanRepasig17(PofansevCowievi

IN

Every year approximately 1.7 million people in the United States sustain a traumatic braininjpobescirecthractbrarec

M.E.H. Pate Rehabilitation, 2655 Villa CreekDrive, Suite 140, Dallas, TX 74254. Addresscorrespondence to M.E.H.; e-mail: mehi@swbell.netDisclosures: board membership, founder, di-rect funds received, Pate Rehabilitation; con-sultancy, direct funds received, Pate Rehabil-itation

P.P. Clinical Research and Development,Pate Rehabilitation, Dallas TXDisclosures: board membership, chairman ofthe board, direct funds received, Pate Reha-bilitation

K.B. Program Development, Pate Rehabilita-tion, Dallas TXDisclosure: nothing to disclose

K.K. Department of Physical Medicine andTexas Southwest-TXose

Center at Galves-

ose

py Development,TXose

May 29, 2012;12

PM ion193 013

Prin .005319ury (TBI), ranging from mild concussions to death [1]. Many persons with TBI requirestacute services to improve functional skills that allow greater self-independence andtter quality of life. However, arguments remain that postacute treatment has not beenentifically proved and that the functional improvements are due solely to naturalovery [2,3]. Fortunately, there now is a wealth of information about natural recoveryough studies of animals [4-8] and humans [9-13] with TBI. Degenerative and reparativeivities begin immediately with cellular and molecular changes that protect an injuredin [4]. These multiple natural mechanisms for recovery include remapping, reweighting,onnecting, and limited regeneration [14]. There also is increasing knowledge about the

Rehabilitation, University ofern Medical Center, Dallas,Disclosure: nothing to discl

B.M. Transitional Learnington, Galveston, TXDisclosure: nothing to discl

D.Q. Research and TheraPate Rehabilitation, DallasDisclosure: nothing to discl

Submitted for publicationaccepted December 14, 20

&R 2013 by the American Academy of Physical Medicine and Rehabilitat4-1482/13/$36.00 Vol. 5, 319-327, April 2

ted in U.S.A. http://dx.doi.org/10.1016/j.pmrj.2012.12eatment Effect Versus Pretreatment Recoveersons With Traumatic Brain Injury: A Studyegarding the Effectiveness of Postacuteehabilitationary Ellen Hayden, PhD, ABPP, Patrick Plenger, PhD, ABPP, Kier Bisonren Kowalske, MD, Brent Masel, MD, Devin Qualls, MS

jective: To evaluate functional improvement following a traumatic brain injury (TBI)er admission to a postacute treatment facility, focusing on the time since injury andalysis of recovery by degree of impairment at admission.sign: A retrospective study of patients who received treatment at a postacute rehabili-ion facility.tting: Postacute rehabilitation for persons with acquired brain injury that involvednsdisciplinary teams.tients: Patients (n 1274) were admitted for treatment less than 5 years after TBI andre assessed on our outcome measures at least 3 times. The patients were then grouped bytime since injury and the severity of impairment at admission.

ethods: Patients received comprehensive multidisciplinary treatment 5 days per week,ours per day.ain Outcome Measurements: Function was assessed by using the Pate Environ-ntally Relevant Program Outcome System (PERPOS) scale at admission, discharge, andproximately every 2 weeks during treatment. By using these assessment scores, the rated degree of improvement were monitored.sults: Postacute rehabilitation yielded significant gains in functioning, with 69% of alltients who demonstrated clinically meaningful gains. The time since injury had anificant impact on gains made in rehabilitation (Ftime time-since-treatment group interaction.75; P .001), with the 0-3 months post injury group outperforming each other group .001 for each comparison). This effect was statistically significant (P .001) for eachthe 3 severity-at-intake subgroups analyzed but was stronger for the severe (F314 9.05)d moderate-to-severe (F425 7.32) than for the mild-to-moderate (F533 2.95)erity-at-intake groups.nclusions: Postacute rehabilitation is associated with functional gains for individuals

th TBI beyond what can be explained by undirected recovery. These findings providedence for postacute rehabilitation as effective care after TBI.

PM R 2013;5:319-327

TRODUCTIONin

D,

im[4,clu

costasuactalsaftboconabrativtocauwiinden

deplaoubywiplemelecaulosproThcliwi

chchbyueindassfurmi[28stuch[30

thahacoforron

leamepronaincshthi

aftinenoutioind6me(mande

funthe[39bil67inledstuwo

injfrosevplibylimnacluwhlon

funrorinjimusmeassmoinwareh

320 Hayden et al TREATMENT EFFECT VS PRETREATMENT RECOVERY IN TBIpact of the external factors, such as types of environment15-17], treatments [18-20], and behaviors [8,21-23], in-ding time frames.Levin et al [24] point out that there has been somenfusion regarding the term recovery. These investigatorste that recovery involves restitution such that the neuralbstrate that mediates the skill or behavior resumes itsivity in a manner similar to that before the injury. Theyo point out that additional mechanisms for improvementer brain injury include compensation, which occurs atth a neural level and a behavioral level. Furthering thenfusion regarding the term recovery is the use of the termtural recovery. This presumes that the organism, and thein, will recover on its own in an adaptive manner irrespec-e of external factors. This has been a problem with regardlimiting funding for rehabilitation after brain injury be-se it is assumed that individuals will recover naturally,thout any special attention to the environment. Clearly, theividual is in an environment after brain injury, and thisvironment can either be adaptive or maladaptive.Given the noted confusion, we would like to create a newfinition of natural recovery as an interaction of neuro-sticity within each environment. More precisely, it is thetcome of how a brain recovers after injury as determinedenvironmental events as well as the natural interactionthin the brain, particularly early after the injury. An exam-of this relationship between neuroplasticity and environ-nt is demonstrated by the effects of neuron shock andarned nonuse [25]. Monkeys that have impaired limbssed by a cortical stroke can be prevented from progressives by using those limbs at least a few hours a day. Thiscess is seen only within the first 2-6 months after injury.e concept of retarding progressive loss of function is nownically used in constraint induced movement therapyth humans [26].Other research on the neuroplasticity and environmentalanges seen in the brain shows that the brain is capable ofanging throughout the life span [22]. Specifically, researchEriksson et al [27] indicates that the dentate gyrus contin-s to acquire new neurons well into adulthood. Althoughividuals may show an abbreviated period of improvementociated with natural recovery, there is clearly a capacity forther neuronal enhancement, which is most likely maxi-zed through appropriate environmental interactions,29]. More recently, neural plasticity has been rigorouslydied with various imaging techniques that looked atanges in the brain over extended courses of treatment-33].It is clear from both human studies and animal researcht cortical remapping begins almost immediately with be-vioral experiences, for example, sensory deprivation. Thisrtical remapping occurs only when the behaviors are rein-

ced [34]. Therefore, there is evidence that specific envi-mental modifications with appropriate reinforcement can

irrsizd to brain-related changes. This interaction of environ-nt, behavior, and brain-related changes (eg, remapping)vides a framework for rehabilitation after TBI. Unfortu-tely, assigning persons with TBI to a nontreatment group isreasingly being viewed as unethical, with more studiesowing that postacute rehabilitation improves outcomes ins patient population [35].Currently, researchers are examining outcomes in patientser acquired brain injury as a function of time since injury,an attempt to account for natural recovery [36]. To thisd, Micklewright et al [37] and High et al [38] reported ontcomes for individuals who attended postacute rehabilita-n at various points in time after injury. Both studiesicated that individuals who entered rehabilitation beforemonths after injury demonstrated the greatest improve-nt. However, individuals who entered the program laterore than 6months for the study byMicklewright et al [37],d more than 1 year for the study by High et al [38])monstrated functional gains as well.Another way to control for the natural recovery of brainctions after TBI without using a control group is to varyintensity of the treatment given. A study by Slade et al] compared standard rehabilitation versus intensive reha-itation. The intensive rehabilitation group was to receive%more therapy (physiotherapy and occupational therapy)a given week. In their study, high-intensity rehabilitationto a significant reduction in length of stay. Results of thesedies suggest treatment effects above and beyond whatuld be expected from undirected natural recovery.A review of the literature on rehabilitation after brainury concluded that there was strong evidence for benefitm formal intervention for individuals with moderate toere injury [40]. These investigators defined multidisci-nary rehabilitation as any intervention that was provided2 or more disciplines in a coordinated effort. There wasited research for specialty programs that involved coordi-ted multidisciplinary rehabilitation. Further questions in-ded which treatment works best with which patients, andich model of service provides the best outcomes over theg term for these patients.To add to the body of research, this study evaluatedctional improvement after admission to a postacute neu-ehabilitation program, which focused on the time sinceury. This study also broke down recovery by degree ofpairment at admission due to a TBI. The present studyed a larger sample of individuals than previously docu-nted. In addition, individuals in this study had beenessed multiple times throughout treatment to allow are detailed analysis of improvement as a function of timetherapy [41,42]. With regard to the current study, first its hypothesized that individuals who received postacuteabilitation would demonstrate functional improvement

espective of the time since injury; second, it was hypothe-ed that individuals who were admitted closer to the time

sinme

M

Pr

ThponespgraseranmomeweouresmeweabpaoggraFa

Sythe(rasuinAscomeThscotalemforlow[Dindvir2totenthefamAllduateab0.7

MadecowhexysianPEverlish

ofMPingansustrstucodicwi

adbilsuanPEantresucuthaatcli20ch

firad6,PEmscomotiothe

theacuthato

321PM&R Vol. 5, Iss. 4, 2013ce injury would demonstrate the best functional improve-nt.

ETHODS

ocedures

e patients who were included in this study participated instacute rehabilitation that involved physiatry, nursing,uropsychology, physical therapy, occupational therapy,eech therapy, and case management coordinated by pro-m managers into transdisciplinary teams. Other specialtyvices available to patients were psychiatry, biofeedback,imal-assisted therapy, residential services, counseling, andre as needed. The program provided comprehensive treat-nt coordinated across professional disciplines 5 days perek, 6 hours per day. The patients were served as eithertpatients or transitional patients. The transitional patientsided in our transitional living facilities but received treat-nt in the same manner as the outpatients. These patientsre graduated to traditional outpatient as they showed theility to live in the less-structured environment. The princi-l program coordinators have doctorates in neuropsychol-y. The medical director is a physiatrist (K.K.). The pro-ms are Commission on Accreditation of Rehabilitationcilities (CARF) accredited.The Pate Environmentally Relevant Program Outcomestem (PERPOS) assesses 3 separate dimensions, includingpatients overall ability level across multiple domains

nge, 1-7), a measure of environmental distraction (mea-red on a 4-point scale), and ameasure of structure inherentthe environment or activity (measured on a 4-point scale).previously published, the environmental variables are

nsidered to be the most critical part in the treatment andasurement model, regardless of the task involved [43].erefore, a composite score is computed by adding there from the overall functional level with the environmen-measures, which are multiplied by a factor of 2 so as tophasize the importance of environmental factors in per-mance. Because the lowest score is 1 in each domain, theest attainable score is 5 ([Function score of 1] 2 istraction score of 1 Structure score of 1]), whichicates total dependence in a quiet, highly structured en-onment. The highest score is 23 ([Function score of 7] [Distraction score of 4 Structure score of 4]), which is

al independence in a highly distractible, low-structuredvironment. The PERPOS is computed at admission andn every 2 weeks thereafter during patient conferences andily conferences. Scores are generated by team consensus.the staff members are trained on the use of the PERPOSring their first 3 months of employment by their immedi-mentor. Interrater reliability has been found to be accept-

le across staff member ratings, with the (average rho (16)7; P .001). The correlation of the PERPOS with the

aftmoyo-Portland Adaptability Inventory (MPAI-4) has beenmonstrated to be excellent (r 0.872; P .001). Thisrrelation enables cross referencing of the data, particularlyen considering generalized guidelines or large populationpectations. The correlation was based on regression anal-s of 844 common data points taken at patient admissiond discharge. Correlating the mean MPAI-4 score at eachRPOS score, with the mean of each PERPOS score con-ted to an MPAI-4 score, resulted in an r 0.996 (unpub-ed data).The MPAI-4 has been demonstrated to be a valid measureoutcome and is currently in its fourth edition [44]. TheAI-4 has been used to evaluate separate areas of function-after acquired brain injury, including ability, adjustment,

d participation. In its current form, the MPAI-4 has mea-res that have been highly developed and have demon-ated well-documented psychometric properties [45]. Pastdies of the MPAI-4 have shown construct validity [46],ncurrent validity with other accepted measures [47], pre-tive validity [46], and, more recently, concurrent validityth PERPOS.All patients (n 1274) with complete data, who weremitted less than 5 years after injury to postacute reha-itation, and had staff assessments on our outcome mea-re documented at least 3 times, were included in thealysis. Each patient was assessed at admission on theRPOS scale to determine that patients impairment leveld was reassessed approximately every 2 weeks unlessatment was interrupted by medical or attendance is-es. In cases in which a treatment gap of 5 weeks oc-rred, the last assessment before the gap was treated ast patients discharge score. A final assessment was donedischarge. Assessment was made by group input in anical staffing according to the PERPOS protocol. Since05, patients are also assessed at admission and dis-arge on the MPAI-4 scale.The 1274 patients assessed in this study were groupedst by severity level based on their PERPOS score atmission to rehabilitation as follows: PERPOS scores of 5,7, or 8 were classified as severe impairment level;RPOS scores of 9, 10, 11, or 12 were classified asoderate-to-severe impairment level; and PERPOSres of 13, 14, 15, 16, or 17 were classified as mild-to-derate impairment level (Table 1). These classifica-ns are based on the level of limitations as identified inMPAI-4 descriptions (Table 2) [45].At each impairment level, the patients were grouped bylength of time between injury and admission into post-te rehabilitation. The time since injury groups were lessn 3months after injury, 3-6months after injury, 6months1 year after injury, 1-2 years after injury, and 2-5 years

er injury. All personal identifying information was re-ved and excluded from this study.

RE

ODulen3-7procuPEpain

rehwhcorateryt12lartre

maatthalevmoPErelit w

Ta

MilMoSeOv

PER

Ta

PESc

evereall Aintendistra

oderassistmodtoler

upervsupelow i

TBIact*U

322 Hayden et al TREATMENT EFFECT VS PRETREATMENT RECOVERY IN TBISULTS

utcome Measuree to the retrospective nature of the present study, thegth of treatment was not constant across patients (range,4 weeks). To control for this confound, we used rate ofgress as our primary dependent measure, which we cal-lated by dividing the admission-to-discharge changes inRPOS by the number of weeks in treatment for eachtient. Thus, the progress rate represents the mean increasePERPOS per week.To test hypothesis 1, namely, that patients who receivedabilitative services would demonstrate improvementile in rehabilitation, we conducted a 1-sample t-test thatmpared actual progress rates to a null rate of zero. Progresses were significantly higher than the pretreatment recov-rate (M [the sample mean of the progress rates] 0.42,

73 43.7; P .001). Effect size analysis revealed a veryge effect size [48] for rehabilitation compared with pre-atment recovery rate (d 1.24). We then tested whether

ble 1. Patient demographics

Level ofImpairment n

PERPOS atAdmission n

d to moderate 533 13, 14, 15, 16, 17 393derate to severe 425 9, 10, 11, or 12 323vere 316 5, 6, 7, or 8 238erall totals 1274 5 to 17 954

POS Pate Environmentally Relevant Program Outcome System.

ble 2. Classifications of TBI severity

RPOSore

MPAI-4 Score (CalculatedApproximate Equivalent)* Severity

5 83-111

Severe

S6 78-827 74-778 70-73

9 66-69Moderate tosevere

M10 61-6511 57-6012 53-56

13 49-52

Mild to moderate

S14 45-4815 40-4416 36-3917 32-35

18 28-31Mild

I19 24-2720 19-23

21 15-18Relatively stable

M22 11-1423 0-10

traumatic brain injury; PERPOS Pate Environmentally Relevant Prog

ivities of daily living; PTA post-traumatic amnesia.npublished data, 2012.not progress rates would vary significantly from the nulle (ie, zero) within each time-since-injury group, by con-cting 5 separate 1-sample t-tests. A significant effect wasnd within each group (0-3 months: t711 40.0, P .001;months: t191 16.4, P .001; 6-12months: t135 11.4,.001; 1-2 years: t136 12.7, P .001; 2-5 years: t96

.9, P .001).The percentage of patients who demonstrated clinicallyaningful gains (defined as a gain of 4 PERPOS) areown in Table 3. Overall, 69% of patients (880/1274)monstrated clinically meaningful gains. At each level oferity at admission examined, more than 50% of patientsde clinically meaningful gains. When the average PERPOSdischarge is compared with data in Table 2, it can be seent many patients made gains of 2 impairment severityels. It is also important to note that, in the mild-to-derate group, all time-since-injury groups have an averageRPOS at discharge of 20 or more, which represents theatively stable impairment level. As stated in our objective,as hypothesized that individuals who received postacute

Men Women

Average Age, y n % Average Age, y

35.9 140 26.3 34.736.4 102 24.0 35.938.7 78 24.7 35.936.8 320 25.1 35.3

Limitations

limitations (physically, cognitively, and/or behaviorally) forDLs; may still be in PTA; requires 24-h supervision; highsity of care; requires very high structure; tolerates very lowctions

ate-to-minimal limitations with basic ADLs; significantance for higher ADLs; frequent-occasional supervision;erate intensity of care; requires moderate structure;ates moderate distractions

ision to independent for basic ADLs; minimal assistance torvision required for higher ADLs; occasional supervision;ntensity of care; not driving, but can use publicportation; modified work; requires less structure; toleratesr distractions

ndence for basic ADLs and supervision for higher ADLs.

e able to live alone or with distant supervision fromgivers; most are able to drive; most are able to work or gohool; requires low structure; tolerates high distractions

tcome System; MPAI-4 Mayo-Portland Adaptability Inventory; ADL transhighe

ndepe

ay bcareto sc

ram Ouorratdufou3-6P 10

meshdesev

%

73.776.075.374.9

rehgarinare

theofinjsigressiginjdifsigTh

disob

acrthemotreintsco1).difinjsigwigrohyacrinjindjur

Ta

Levels

-12: Mo

TimI

WithRPOSin >4

3 0.03-6 5.06 m 8.51-2 0.52-5 1.5Tota 9.9

PER ischarg

FigwitaftOumeprolineeamo75%latco

323PM&R Vol. 5, Iss. 4, 2013abilitation will demonstrate functional improvement re-dless of time since injury. Such improvement is indicatedTable 3 and in Figures 1, 2, and 3. PERPOS gains per weekshown graphically as the slopes of the trend lines.To test hypothesis 2, that time since injury would affectefficacy of rehabilitation, we performed a 1-way analysisvariance that compared progress rates across time-since-ury groups. Consistent with our hypothesis, we found anificant overall effect of time-since-injury group on prog-s rate (F 53.0; P .001). Post hoc Tukey tests revealednificantly greater progress rates for the 0-3 months postury group compared with each other group (range of meanferences, 0.22-0.30; P .001 for each comparison). Nonificant differences were seen among any other groups.e mean progress rates for each time-since-injury group are

ble 3. Clinically significant improvements

% Patients Progressing > PERPOS

PERPOS 5-8: Severe PERPOS 9

e Afternjury

No. WithGain >4

% WithPERPOSGain >4

AveragePERPOS atAdmission

AveragePERPOSat DC

No. WithGain >4

%PEGa

mo 116 of 155 74.8 6.59 15.05 204 of 255 8mo 32 of 70 45.7 6.09 13.97 27 of 60 4o to 1 y 5 of 31 16.1 7.16 13.80 24 of 41 5y 16 of 43 32.7 6.60 12.56 26 of 43 6y 3 of 17 17.6 6.41 11.33 16 of 26 6l 172 of 316 54.4 297 of 425 6

POS Pate Environmentally Relevant Program Outcome System; DC d

ure 1. Rate of recovery over time. Includes 316 patientsh traumatic brain injury who were admitted less than 5 yearser injury at the Pate Environmentally Relevant Programtcome System (PERPOS) score of 5 to 8. Each line shows thean patient progress while 75% of patients remained in thegram. For Figures 1-3, the data were plotted and the trends were computed to the point where 75% of the patients inch group remained in the program. Patients dropped out atst weekly intervals. Continuing to average data beyond thepoint left a group less representative of the whole popu-

ion. The majority of patients are accounted for by this

nservative cutoff. paplayed in Figure 4 and highlights significant differencesserved.Hypothesis 2 was evaluated by using weekly PERPOSoss the first 9 weeks of treatment. Weeks 1-9 were set astime frame for this analysis. This time frame representedst intervals during which 75% of patients remained inatment. To control for confounding effects of severity atake, we transformed weekly PERPOS into differenceres relative to week 1 (ie, PERPOSweek n PERPOSweekThen we ran an analysis of variance, with these PERPOSference scores as the dependent factor and time-post-ury group as a between-subjects factor. This revealed anificantmain effect for time (F 500, P .001, consistentth hypothesis 1) and a significant time time-since-injuryup interaction (F 17.75, P .001, consistent withpothesis 2). The mean improvements in PERPOS scoresoss the first 9 weeks of rehabilitation for each time-post-ury group are shown in Figure 5. Post hoc Tukey testsicated greater improvement for the 0-3 months post in-y group than for each of the other groups (mean difference

and Their Average Discharge Outcomes

derate to Severe PERPOS 13-17: Mild to Moderate

AveragePERPOS atAdmission

AveragePERPOSat DC

No. WithGain >4

% WithPERPOSGain >4

AveragePERPOS atAdmission

AveragePERPOSat DC

10.84 18.75 224 of 305 73.4 14.54 20.9710.70 17.15 38 of 60 63.3 14.75 20.6610.61 17.67 41 of 62 66.1 14.97 20.5410.79 17.46 22 of 52 42.3 14.87 20.5010.85 18.31 26 of 54 48.1 14.69 20.35

351 of 533 65.9

e.

ure 2. The rate of recovery over time, which includes 425tients with traumatic brain injury (TBI) who were admitteds than 5 years after injury, at the Pate Environmentallylevant Program Outcome System (PERPOS) score of 9-12.ch line shows the mean patient progress while 75% ofFigpalesReEatients remained in the program.

ransiggroco

whpointvarwean

Theacefftimacrnosevmi

DI

Restrregsingrateddepasevwhfun4 Pageitashthe

FigpalesReEapa

Figinju

Taan

In

MilMoSe

324 Hayden et al TREATMENT EFFECT VS PRETREATMENT RECOVERY IN TBIge, 0.68-1.05, P .001 for all comparisons); again, nonificant differences were seen among any of the otherups (mean difference range, 0.07-0.36; P .30 for eachmparison).To elaborate on the tests of hypothesis 2, it was exploredether or not longer time-since-injury group predictedorer treatment response across different levels of severity atake. This was done by conducting a separate analysis ofiance for each severity-at-admission group, enteringekly PERPOS difference scores as the dependent factord time-since-injury group as a between-subjects factor.

ure 3. The rate of recovery over time, which includes 533tients with traumatic brain injury (TBI) who were admitteds than 5 years after injury, at the Pate Environmentallylevant Program Outcome System (PERPOS) score of 13-17.ch line shows the mean patient progress while 75% oftients remained in the program.

ure 4. The mean progress rates for different time-since-

ry groups.

PERe results of these analyses are shown in Table 4. Withinh admission severity group, there was a significant mainect of time-since-injury group as well as a significant timee-since-injury group interaction. Although significantoss all 3 levels of intake severity, this interaction wastably larger for the severe (F 9.05) and moderate-to-ere (F 7.32) admission severity groups than for theld-to-moderate severity group (F 2.95).

SCUSSION

sults of this study indicate that patients with TBI demon-ated improved functioning after postacute rehabilitation,ardless of severity of impairment at admission or timece injury to admission to a postacute rehabilitation pro-m. The rate of improvement was greater for those admit-within 3 months of their injury. Although all the patients

monstrated functional gains, irrespective of degree of im-irment at admission, it was found that individuals withere levels of impairment demonstrated less improvementen admitted later in time after injury. Clinically relevantctional gains (ie, change in impairment severity of at leastERPOS points) of treatment were seen for a large percent-(69%) of all the patients admitted into postacute rehabil-

tion. When compared with the data in Table 2, it wasown that, when patients achieved a change of 4 points onPERPOS, they generally have moved from one limitation

Figure 5. Gains over time by the time-since-injury group.

ble 4. Results of analysis of variance testing effects of timed severity-at-intake group on PERPOS

take Severity

Main Effect(time)

Interaction(time

severity-at-intake group)

F P F P

d-moderate 248 .001 2.95 .001derate-severe 180 .001 7.32 .001vere 68 .001 9.05 .001POS Pate Environmentally Relevant Program Outcome System.

levanthetio

repByvalmaocposiothoofwhtudanpreati

nefindiscoitytheindtimsigfoutiasusigonindma2-5ind

sigrestreanpovidquothadticforshciaena h

bilPEmamePEsomallthe

ofpowilearecneeryabtheprolearehdigratdevarinjmaanmomoproyea(annoprosutionewhcan

C

ThassbeThrehemintlimtieme

325PM&R Vol. 5, Iss. 4, 2013el to another. Such changes in levels of limitation indicateimproved ability on functionally relevant tasks, such asability to stay home alone and to use public transporta-

n.The present study also demonstrates the utility of usingeated measures in clinical research with this population.using frequent (ie, every 2 weeks) assessment with a singleidated measure, it is possible to detail the rate of progressde by each patient and to empirically verify that progresscurred in a linear fashion. This would not have beenssible, had data been collected only at the time of admis-n and at the time of discharge. Because neuroplasticity isught to affect both the rate of learning as well as the limitswhat can be learned, it is critical to evaluate the rate atich gains are made as opposed to their absolute magni-e. It is suggested that using frequent repeated measuresd evaluating change throughout treatment, as done in thesent study, provides a useful means of empirically evalu-ng the efficacy of various rehabilitative techniques.The present findings have clear implications for providingurorehabilitative services in clinical practice. First, thedings indicate that efforts should be made to begin trans-ciplinary rehabilitation within 3 months of injury. This isnsistent with prior research that showed that neuroplastic-is greatest immediately after an injury and decreases withpassage of time [37,38]. More importantly, these findingsicate that decreases in neuroplasticity with increasinge since injury do not preclude patients from makingnificant gains many months or years after injury. It wasnd that patients 2-5 years after injury still made substan-l gains in functioning, which calls into question the as-mption that these patients are no longer capable of makingnificant improvements in functioning. These findings leade to question the argument for denying treatment to theseividuals on the grounds that it is too late for progress to bede. Indeed, results of this study suggest that individualsyears after injury show about the same rate of progress asividuals as little as 3-6 months after injury.Due to the archival nature of this research, there arenificant limitations that should be addressed in futureearch. One limitation is that there was not a true no-atment control group with which to compare the degreed rate of improvement noted for individuals who receivedstacute rehabilitation. Future studies could monitor indi-uals who are on a wait list due to insurance reasons, toantify improvement without specific rehabilitation. An-er limitationwas that data regarding patient history beforemission to postacute rehabilitation (eg, injury characteris-s, amount and type of rehabilitative services received be-e admission) was incomplete. Future research effortsould systematically evaluate injury characteristics, espe-lly the length of posttraumatic amnesia, in addition to

vironmental factors such as any treatment received both inospital and at home before admission to postacute reha-

poyeaitation. A final identified limitation relates to the fact thatRPOS is first scored on admission to this facility, and, inny cases, it is impossible to obtain any previous measure-nts from the time of injury to admission. Thus, admissionRPOS scores after 0-3 months after injury likely includee achieved improvement. Future statistical models may

owmore information by using days from injury along withother variables with these data.The present study promotes a discussion of the questionbenefit and/or impact from rehabilitation (specifically,stacute rehabilitation) on accelerating functional gainsth individuals who were brain injured. This discussion willd to more detailed analyses of both the process of naturalovery and the specific environmental factors that affecturoplastic mechanisms and, ultimately, maximize recov-rates. We postulate that natural recovery, as defined

ove, is not a uniform process at all and is better viewed aslikely product of physiological readiness, neuroplasticcesses, and unidentified, unspecified, and unstructuredrning contingencies that range across hospital, home, andabilitation environs. The importance of learning para-ms (specifically, reinforcement rates) in maximizing thee and extent of neuroplastic changes have been clearlymonstrated [49,50]. Therefore, natural recovery willy across individuals not only by type and/or extent ofury but also by timing and salience of environmentalnipulation. The current study supports previous researchd leads us to ask not whether rehabilitation works, but,re importantly, which rehabilitation interventions mightst effectively take advantage of the neuronal recoverycess. The fact that rehabilitation efforts with patients 2-5rs after brain injury produce significant treatment gainsd reduced functional limitations), clearly supports thetion that some structured environmental intervention isductive beyond naturally occurring contingencies. It is

ggested that the current study, which supports rehabilita-n effectiveness, in conjunction with a growing body ofuroplasticity research, argues for the future study not ofether rehabilitation works but,rather, how rehabilitationwork more efficiently and effectively.

ONCLUSION

e present study indicated that postacute rehabilitation isociated with functional gains for individuals with TBIyond what can be expected by undirected recovery alone.e findings of the study provide evidence for postacuteabilitation as the standard of care after TBI, with anphasis on early entry after injury. As expected, earlyervention produced the greatest reduction in functionalitations. This finding is especially important for the pa-nts with the most severe limitations. Regardless of impair-nt, there is evidence that supports the effectiveness of

stacute rehabilitation for individuals, even those severalrs after injury.

AC

Wrevingtoounoidepri

RE1.

2.

3.

4.

5.

6.

7.

8.

9.

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

326 Hayden et al TREATMENT EFFECT VS PRETREATMENT RECOVERY IN TBIKNOWLEDGMENTS

e thank Rich Capriotti, PhD, for editorial assistance inising this manuscript, and Matthew Capriotti for provid-statistical analysis of the data. Additionally, we would likethank Mr. Henry Irving for his extensive work on thetcome database. Institutional review board approval wast required for the scope of this study because all patientntifying material was removed before data analysis. Thenciples of the Declaration of Helsinki were followed.

FERENCESFaul M, Xu L, Wald MM, Coronado VG. Traumatic brain injury in theUnited States: Emergency department visits, hospitalizations, anddeaths. Atlanta, GA: Centers for Disease Control and Prevention, Na-tional Center for Injury Prevention and Control; 2010.Davis AE. Cognitive impairments following traumatic brain injury:Etiologies and interventions. Crit Care Nurs Clin North Am 2000;12:447-456.Ratan RR, Noble M. Novel multi-modal strategies to promote brain andspinal cord injury recovery. Stroke 2009;40(Suppl):S130-S132.Kolb B, Cioe J, Williams P. Neuronal organization and change afterbrain injury. In: Raskin SA, ed. Neuroplasticity and Rehabilitation. NewYork, NY: Guilford Press; 2011, 13-37.Buonomano J, Merzenich MM. Cortical plasticity: From synapses tomaps. Annu Rev Neurosci 1998;21:149-186.Friedman RM, Chen LM, Roe RW. Modality maps within primatesomatosensory cortex. Proc Natl Acad Sci U S A 2004;101:1272-1279.Nudo RJ, Bury S. Motor and sensory reorganization in primates. In:Raskin SA, ed. Neuroplasticity and Rehabilitation. New York, NY:Guilford Press; 2011, 65-102.Jones TA. Experience-dependent changes in nonhumans. In: RaskinSA, ed. Neuroplasticity and Rehabilitation. New York, NY: GuilfordPress; 2011, 38-64.Morris DM, Bickel CS. Activity-based interventions for neurorehabili-tation. In: Raskin SA, ed. Neuroplasticity and Rehabilitation. New York,NY: Guilford Press; 2011, 117-148.

. Kleim JA, Jones TA. Principles of experience dependent neural plastic-ity: Implications for rehabilitation after brain damage. J Speech LangHear Res 2008;51:S225-S239.

. Rijntjes M. Mechanisms of recovery in stroke patients with hemiparesisor aphasia: New insights, old questions and the meaning of therapies.Curr Opin Neurol 2006;19:76-83.

. Saur D, Lange R, Baumagaertner A, et al. Dynamics of language reor-ganization after stroke. Brain 2006;129:1371-1384.

. Lillie RA, Mateer CA. Neuronal organization and change after braininjury. In: Raskin SA, ed. Neuroplasticity and Rehabilitation. NewYork, NY: Guilford Press; 2011, 257-284.

. Seung S. Connectome: How the Brains Wiring Makes UsWhoWe Are.New York, NY: Houghton Mifflin Harcourt; 2012.

. Biernaskie J, Corbett D. Enriched rehabilitative training promotesimproved forelimb motor function and enhanced dendritic growthafter focal ischemic injury. J Neurosci 2001;21:5272-5280.

. Johansson BB. Functional outcome in rats transferred to an enriched environ-ment 15 days after focal brain ischemia. Stroke 1996;27:324-326.

. Will B, Kelche C. Environmental approaches to recovery of functionfrom brain damage: A review of animal studies (1981 to 1991). In: RoseFD, Johnson DA, eds. Recovery From Brain Damage: Reflections andDirections. New York, NY: Plenum Press; 1992, 799-804.. Johansson BB. Brain plasticity and stroke rehabilitation: The WillisLecture. Stroke 2000;223-230.. Kleim JA, Jones TA, Shallert T. Motor enrichment and the induction ofplasticity before or after brain injury. NeurochemRes 2003;28:1757-1769.

. Will B, Galani R, Kelche C, RosenzweigMR. Recovery from brain injury inanimals: Relative efficacy of environmental enrichment, physical exerciseor formal training (1990-2002). Prog Neurobiol 2004;72:167-182.

. de Witt BW, Ehrenberg KM, McAloon RL, et al. Abbreviated environ-mental enrichment enhances neurobehavioral recovery comparably tocontinuous exposure after traumatic brain injury. Neurorehabil NeuralRepair 2011;25:343-350.

. Schallert T, Leasure L, Kolb B. Experience-associated structural events,subependymal cellular proliferative activity, and functional recoveryafter injury to the central nervous system. J Cereb Blood Flow Metab2000;20:1513-1528.

. Carmichael ST. Cellular and molecular mechanisms of neural repairafter stroke: Making waves. Ann Neurol 2006;59:735-742.

. Levin MF, Kleim JA,Wolf SL.What do motor recovery and compen-sation mean in patients following stroke? Neurorehabil Neural Repair2009;23:313-319.

. Taub E, Crago JE, Burgio LD, et al. An operant approach to rehabilita-tion medicine: Overcoming learned nonuse by shaping. J Exp AnalBehav 1994;61:281-293.

. Wolf SL, Lecraw DE, Barton LA, Jann BB. Forced use of hemiplegicupper extremities to reverse the effect of learned nonuse among chronicstroke and head-injured patients. Exp Neurol 1989;104:125-132.

. Eriksson PS, Perfilieva E, Bjrk-Eriksson T, et al. Neurogenesis in theadult human hippocampus. Nat Med 1998;4:1313-1317.

. GrayDS. Slow-to-recover severe traumatic brain injury: A reviewof outcomesand rehabilitation effectiveness. Brain Inj 2000;14:1003-1014.

. Kliem JA. Neural plasticity and neurorehabilitation: Teaching the newbrain old tricks. J Commun Disord 2011;44:521-528.

. Nakamura T, Hillary FG, Biswal BB. Resting network plasticity follow-ing brain injury. PLoS One 2009;4:e8220.

. Muoz-Cespedes JM, Rios-Lago M, Paul N, Maestu F. Functionalneuroimaging studies of cognitive recovery after acquired brain damagein adults. Neuropsychol Rev 2005;15:169-183.

. Zorowitz R, Brainin M. Advances in brain recovery and rehabilitation2010. Stroke 2011;42:294-297.

. Chen H, Epstein J, Stern E. Neural plasticity after acquired brain injury:Evidence from functional neuroimaging. PMR 2010;2(Suppl 2):S306-S312.

. Wittenburg GF. Experience, cortical remapping, and recovery in braindisease. Neurobiol Dis 2010;37:252-258.

. Turner-Stokes L. The evidence for the cost-effectiveness of rehabilita-tion following acquired brain injury. Clin Med 2004;4:10-12.

. Carod-Artal FJ, Medeiros MS, Horan TA, Braga LW. Predictive factorsof functional gain in long-term stroke survivors admitted to a rehabil-itation programme. Brain Inj 2005;19:667-673.

. Micklewright JL, Yutsis M, Smigielski JS, Brown AW, Bergquist TF.Point of entry and functional outcomes after comprehensive day treat-ment participation. Arch Phys Med Rehabil 2011;92:1974-1978.

. HighWM Jr, Roebuck-Spencer T, Sander AM, Struchen MA, Sherer M.Early versus later admission to postacute rehabilitation: Impact onfunctional outcome after traumatic brain injury. Arch Phys Med Reha-bil 2006;87:334-342.

. Slade A, Tennant A, Chamberlain MA. A randomised controlled trial todetermine theeffectof intensityof therapyupon lengthof stay inaneurologicalrehabilitation setting. J Rehabil Med 2002;34:260-266.

. Turner-Stokes L, Disler PB, Nair A, Wade DT. Multi-disciplinary reha-bilitation for acquired brain injury in adults of working age. CochraneDatabase Syst Rev 2005;(3):CD004170.

. Cappa KA, Conger JC, Conger AJ. Injury severity and outcome: Ameta-analysis of prospective studies on TBI outcome. Health Psychol2011;30:542-560.

. Waehrens EE, Fisher AG. Improving quality of ADL performance after

rehabilitation among people with acquired brain injury. Scand J OccupTher 2007;14:250-257.

43

44

45

46

47

CMIndtra

a.b.c.d.

An

327PM&R Vol. 5, Iss. 4, 2013. Hayden ME, Moreault AM, LeBlanc J, Plenger PM. Reducing level ofhandicap in traumatic brain injury: An environmentally basedmodel oftreatment. J Head Trauma Rehabil 2000;15:1000-1021.

. Malec JF, Thompson JM. Relationship of the Mayo-Portland Adaptabil-ity Inventory to functional outcome and cognitive performance mea-sures. J Head Trauma Rehabil 1994;9:1-15.

. Malec J. The Mayo Portland Adaptability Inventory. The Center forOutcome Measurement in Brain Injury. 2005. Available at: http://www.tbims.org/combi/mpai. Accessed August 31, 2012.

. Bowers D, Kofroth L. Comparison: Disability Rating Scale and Func-tional Independence Measure during recovery from traumatic braininjury. Arch Phys Med Rehabil 1989;70:A58.

. Malec JF, Kragness M, Evans RW, Finlay KL, Kent A, Lezak MD.Further psychometric evaluation and revision of the Mayo-PortlandAdaptability Inventory in a national sample. J Head Trauma Rehabil2003;18:479-492.

48. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed.Hillsdale, NJ: Lawrence Erlbaum; 1988.

49. Jenkins WM, Merzenich MM, Ochs MT, Allard T, Guc-Robles E.Functional reorganization of primary somatosensory cortex in adultowl monkeys after behaviorally controlled tactile stimulation. J Neuro-physiol 1990;63:82-104.

50. Wang X, Merzenich MM, Sameshima K, Jenkins WM. Remodeling ofhand representation in adult cortex determined by timing of tactilestimulation. Nature 1995;378:71-75.

This CME activity is designated for 1.0 AMA PRACategory 1 Credit andcan be completed online at me.aapmr.org. Log on to www.me.aapmr.org,go to Lifelong Learning (CME) and select Journal-based CME from thedrop down menu. This activity is FREE to AAPM&R members and $25 fornon-members.

E Questionividuals in which post-injury time frame demonstrated the most significant gains when admitted to post-acute rehabilitation programs afterumatic brain injury with mild to moderate impairment?

0 to 3 months3 to 6 months6 months to 1 year1 year to 2 years

swer online at me.aapmr.org

Treatment Effect Versus Pretreatment Recovery in Persons With Traumatic Brain Injury: A Study Re ...IntroductionMethodsProcedures

ResultsOutcome Measure

DiscussionConclusionAcknowledgmentsReferences