Read without ads and support Scribd by becoming a Scribd Premium Reader.
See Premium Plans
 
CLINICAL INVESTIGATIONS
A Cognitive Training Program Based on Principles of BrainPlasticity: Results from the Improvement in Memory withPlasticity-based Adaptive Cognitive Training (IMPACT) Study
Glenn E. Smith, PhD,
Ã
Patricia Housen, PhD,
w 
Kristine Yaffe, MD,
z 
§ 
k
Ronald Ruff, PhD,
z 
#
Robert F. Kennison, PhD,
z 
ÃÃ
Henry W. Mahncke, PhD,
ww 
and Elizabeth M. Zelinski, PhD
w 
OBJECTIVES:
To investigate the efficacy of a novel brainplasticity–based computerized cognitive training programin older adults and to evaluate the effect on untrained mea-sures of memory and attention and participant-reportedoutcomes.
DESIGN:
Multisite randomized controlled double-blindtrial with two treatment groups.
SETTING:
Communities in northern and southern Cali-fornia and Minnesota.
PARTICIPANTS:
Community-dwelling adults aged 65and older (N
5
487) without a diagnosis of clinically sig-nificant cognitive impairment.
INTERVENTION:
Participantswererandomizedtoreceivea broadly-available brain plasticity–based computerized cog-nitive training program (intervention) or a novelty- and in-tensity-matched general cognitive stimulation programmodeling treatment as usual (active control). Duration of trainingwas1hourperday,5daysperweek,for8weeks,fora total of 40 hours.
MEASUREMENTS:
The primary outcome was a compos-ite score calculated from six subtests of the RepeatableBattery for the Assessment of Neuropsychological Statusthat use the auditory modality (RBANS Auditory Memory/ Attention). Secondary measures were derived from perfor-mance on the experimental program, standardized neuro-psychological assessments of memory and attention, andparticipant-reported outcomes.
RESULTS:
RBANS Auditory Memory/Attention improve-ment was significantly greater (
P
5
.02) in the experimentalgroup (3.9 points, 95% confidence interval (CI)
5
2.7–5.1)than in the control group (1.8 points, 95% CI
5
0.6–3.0).Multiple secondary measures of memory and attentionshowed significantly greater improvements in the experi-mentalgroup(wordlisttotalscore,wordlistdelayedrecall,digits backwards, letter–number sequencing;
P
o
.05), asdid the participant-reported outcome measure (
P
5
.001).No advantage for the experimental group was seen in nar-rative memory.
CONCLUSION:
The experimental program improvedgeneralized measures of memory and attention more thanan active control program.
J Am Geriatr Soc 2009.
Key words: clinical trial; cognitive decline; computerizedcognitive training; participant-reported outcomes; brainplasticity
C
ognitive decline is associated with risk for functionaldecline, nursing home placement, and mortality.
1–3
Inolder individuals, concerns about forgetfulness are wide-spread and are associated with depression and anxiety.
4–6
Interventions that reliably improve cognitive function thushave the opportunity to substantially improve the healthand quality of life of older individuals.Two general approaches for maintaining or improvingcognitive function in older adults have emerged. The firstapproach is focused on direct instruction of putatively use-ful strategies.
7–12
Although improvement on cognitive testsis generally seen after direct strategy instruction, perfor-
Portions of this research were presented as a poster presentation at the60th Annual Scientific Meeting of the Gerontological Society of America,November 16–20, 2007, San Francisco, CA; 36th Annual Scientific Meetingof the International Neuropsychological Society, February 6–9, 2008,Waikoloa, HI; Annual Meeting of the American Academy of Neurology,April 12–19, 2008, Chicago, IL; 2008 American Geriatrics Society AnnualScientificMeeting(Encore),April30–May5,2008,Washington,DC;and6thInternational Conference of the International Society for Gerotechnology, June 4–6, 2008, Pisa, Italy.Addresscorrespondence toGlennE.Smith,MayoClinic,200FirstStreetSW,Rochester, MN 55905. E-mail: smitg@mayo.eduDOI: 10.1111/j.1532-5415.2008.02167.xFrom the
Ã
Department of Psychiatry and Psychology, Mayo Clinic, Roches-ter, Minnesota;
w
Leonard Davis School of Gerontology, University of South-ern California, Los Angeles, California; Departments of 
z
Psychiatry;
§
Neurology; and
k
Epidemiology, University of California, SanFrancisco, San Francisco, California;
#
Division of Physical Medicine andRehabilitation, Stanford University, Stanford, California;
ÃÃ
Department of Psychology, California State University, Los Angeles, California; and
ww
Posit Science Corporation, San Francisco, California.
 JAGS 2009
r
2009, Copyright the Authors Journal compilation
r
2009, The American Geriatrics Society
0002-8614/09/$15.00
 
mance gains typically do not generalize beyond tasks cor-responding directly to the strategies taught,
13–15
and it isnot clear that older adults continue to use learned strategiesovertime.
15
Asaresult,strategytrainingprogramshavenotbeen widely adopted.A second approach is derived from studies in animals
16
and humans
17–20
that suggest that nonspecific cognitivestimulationreducestheriskofcognitivedecline.Thishasledto the practice of encouraging older adults to engage in ev-eryday cognitively stimulating activities,
14,21,22
but the ret-rospective and observational designs of the human studieshave led to difficulty interpreting the direction of causationbetween cognitive function and cognitively stimulatingactivities.
22
Regardless of the design principles, large-scale ran-domized controlled trials of training programs that arebroadly available for patient use are lacking, limiting theability of physicians to make evidence-based recommenda-tions to older adults experiencing cognitive decline.In recent years, recognition of the importance of sen-sorysystemfunctiontocognitivefunctionhaspromptedthedevelopment of a novel approach for treating age-relatedcognitive decline. It has been proposed that age-related re-ductions in the quality of neural information flowingthrough peripheral and central sensory systems to cogni-tivesystems contributetoage-relatedcognitivedecline.
23,24
Animal and human studies have demonstrated that the per-formance of sensory systems in the cerebral cortex can besubstantially improved through intensive learning andpractice and that plastic brain changes across networks of relevantcorticalareasinthecentralnervoussystemmediatetheseimprovements.
25,26
Consequently,acognitivetrainingprogram designed to improve central sensory system func-tion could potentially improve cognitive function in olderadults.
27
Resultsare reported fromtheImprovement in Memorywith Plasticity-based Adaptive Cognitive Training (IM-PACT) study, a large randomized controlled two-arm clin-ical trial using a broadly available cognitive trainingprogram (Brain Fitness Program, Posit Science, San Fran-cisco, CA). The program is designed to improve the func-tion of the auditory system through intensive brainplasticity–based learning and has shown promise insmaller-scale studies.
27,28
The current study builds uponthe earlier studies by broadening the outcome measures toinclude a positive control for task learning, more memoryand attentionmeasures, and participant-reported outcomes(PROs), as well as being powered to detect across groupdifferences.The primary objective was to evaluate the efficacy of this experimental treatment (ET) training program by com-paring the magnitude of improvements on untrained mea-sures of memory and attention between the ET trainingprogram and an active control (AC) training program thatengagedlearningprocessesbutwasnotdesignedtoimproveauditory system function.
METHODSDesign
This was a multisite (Los Angeles, CA; Rochester, MN; SanFrancisco, CA) randomized controlled double-blind trial.
Participants
Inclusion criteria were aged 65 and older, Mini-MentalState Examination (MMSE
29
) score of 26 or greater, En-glish fluency, and ability to make time commitment. Ex-clusion criteria were major neurological or psychiatricillness history, including any history of stroke, transient is-chemic attack, or traumatic brain injury; acetylcholinester-ase inhibitor use; current substance abuse; significantcommunicative impairments; and concurrent enrollmentin other studies. Recruitment took place through advertise-ments, flyers, direct mail, and presentations.
Procedures
Institutionalreview boardapproval and written participantconsent describing the ETand AC programs were obtained.No reimbursement was offered, but computer equipmentwas provided to all participants at no cost during the train-ing period. Interventions were self-administered at partic-ipants’ homes; assessments occurred in clinical offices.Participants completed 40 sessions (1 h/d, 5 d/wk, for 8weeks).Participants notadherent tothetrainingregimen(com-pleting
o
10 sessions in the first month or skipping
4
10consecutivesessionsthereafter),thosewhoduringthestudyno longer met the inclusion and exclusion criteria, or thosevoluntarily withdrawing consent were discontinued fromthe study. In all cases, their pretraining data were retainedfor the intention-to-treat (ITT) analysis.Participants were given sequential study identificationnumbers and randomly assigned to an age-stratified treat-ment group (20% aged 65–69, 40% aged 70–79, 40% aged80). A random sequence of ETand AC assignments withineach age stratum was generated before study commence-ment. Sites requested randomization allocation throughe-mail, and a single staff member fulfilled requests throughconcealed randomization allocation sequence administered.Randomization was blocked according to site and age.During the initial visit, an unblinded trainer installedthe computer and provided individualized instruction andpretraining as needed in the use of the equipment andtraining program for both groups. The trainers used stan-dardized scripts to describe the rationales and benefits of bothprogramstomaintainparticipantblinding.ETandACtraining tasks were self-administered. Trainers contactedparticipants weekly to identify and resolve technical prob-lems and record adverse events.Participants and clinicians administering and scoringoutcome measures were blinded. Effectiveness of blindingwas evaluated by administering a posttraining question-nairetocompareETandACgroupself-reportsofperceivedchange in cognitive function and comparing proportions of ETand AC participants who voluntarily withdrew consent.
Training Programs
 Brain Plasticity–Based ET 
The ET consisted of six computerized exercises designed toimprove the speed and accuracy of auditory informationprocessing. Exercises continuously adjusted difficulty touser performance to maintain an approximately 85% cor-rect rate. Correct trials were rewarded with points and an-imations. Exercises contained stimulus sets spanning the
2
SMITH ET AL.
2009 JAGS
 
acoustic organization of speech. The exercises includedtime order judgment of pairs of frequency-modulatedsweeps, discrimination of confusable syllables, recognitionof sequences of confusable syllables, matching pairs of confusable syllables, reconstruction of sequences of verbalinstructions, and identification of details in a verbally pre-sented story. During the initial stages of training in all exer-cises, all auditory stimuli were processed to exaggerate therapid temporal transitions within the sounds by increasingtheir amplitude and stretching them in time. The goal of theprocessing was to increase the effectiveness by which thesestimuliengageanddriveplasticchangesinbrainsystemsthat,in older adults, exhibit relatively poor temporal responseproperties.
27
This exaggeration was gradually removed overthe course of the training period such that, by the end of training, all auditory stimuli had temporal characteristicsrepresentative of real-world rapid speech. In each trainingsession,aparticipantworkedwithfourofthesixexercisesfor15 minutes per exercise. Adherence was monitored usingelectronic data upload after each training session.
 Educational Training AC
ACtrainingwasrequiredtohavefacevalidity;beconsistentwith common physician recommendations for cognitivestimulation; and match ET for training time, audio-visual presentation, and computer use. Thus the programemployed a learning-based training approach in which par-ticipants used computers to view digital video disc (DVD)-based educational programs on history, art, and literature.Participants answered written quizzes after each trainingsession that required the specific factual content knowledgepresented by the DVD in that session. These quizzes servedto ensure attention and learning during the training sessionand allowed quantitative measurement of compliance.The trial design did not incorporate a no-contact con-trol (NCC) condition based on comparisons of NCC andAC group auditory memory outcomes in previous workthat showed equivalent cognitive improvements in NCCand AC groups.
27,28
Pretraining Characterization Measures
Demographics (age, education, sex, ethnicity, first lan-guage), cognitive status (MMSE, estimated intelligencequotient (Wechsler Test of Adult Reading)
30
), depression(15-item Geriatric Depression Scale score
31
), and sensoryfunctions (audiometric function, tinnitus, hearing aid, eye-glass use) were measured.
Outcome Measures
The primary outcome measure was derived from Repeat-able Battery for the Assessment of NeuropsychologicalStatus (RBANS
32
), a standardized neuropsychological as-sessment battery that is sensitive to mild cognitive deficits.Because the ET focused on improving auditory processing,the primary outcome measure (RBANS Auditory Memory/ Attention) was derived from the six RBANS subtests thatuse orally presented speech as stimuli (list learning, storymemory,digitspanforward,delayedfreelistrecall,delayedlist recognition, delayed free story recall). Raw scores wereconvertedtoscaledscoresbasedon look-uptablesmappingnormative RBANS population data to optimal Gaussiandistributions. Delayed list recall and recognition weresummed before scaling to allow inclusion of the skewedrecognition data. The five equally weighted scaled scoreswere then summed and mapped to a composite index score(average
5
100, standard deviation
Æ
15).BecausetheRBANSmayexhibitceilingeffectsinhighlyfunctioning older adults,
28
additional assessments of audi-tory memory and attention were used to provide furtherinformation about the robustness of generalization. Stan-dardized published measures that met the following criteriawere chosen: use of orally presented speech as stimuli, sen-sitivity to age-related cognitive decline, lack of test–retesteffects through use of multiple forms or stimuli that are notrememberedacrossassessmentvisits,andrelevancetomem-ory and attention. The measures used were Rey AuditoryVerbal Learning Test (RAVLT
33
) total score (sum of trials1–5) and word list delayed recall, Rivermead BehavioralMemory Test (RBMT
34
) immediate and delayed recall, andWechsler Memory Scale (WMS-III
35
) letter-number se-quencing (LNS) and digit span backwards tests. An overallcomposite score (Overall Memory) combining RAVLT totalscore and word list delayed recall, RBMT immediate anddelayedrecall,andLNSanddigitsbackwardswasderivedasdescribed for RBANS Auditory Memory/Attention; becauseof the lack of published co-normed data, standardizationwas based on the pretraining score distribution.Secondary outcomes also included a directly trainedmeasure of exercise performance derived from the ET pro-cessing speed exercise, as well as a pre–post PRO measurethat assesses perceptions of cognitive abilities (CognitiveSelf-Report Questionnaire, CSRQ-25
36
). The CSRQ-25consists of 25 statements about cognition and mood in ev-eryday life over the past 2 weeks, answered using a 5-pointLikert scale. The CSRQ-25 was developed as a PRO, be-cause existing PROs do not include questions relevant tocognitive training, are not appropriate for healthy olderadult population, or were designed to measure cognitiveimpairment rather than be sensitive to improvement. TheCSRQ-25 was validated using factor analysis on 207healthy older adults before this study. Concurrent and di-vergent validity were established by examining correlations(
P
o
.05) with subscales of the Life Satisfaction Scale(LSS),
37
Cognitive Failures Questionnaire (CFQ),
38
andGeriatric Depression Scale.
31
Reliability measures includeCronbach alpha 0.91, Spearman-Brown split-half reliabil-ity 0.94, and 2-month test–retest reliability 0.85.A posttraining questionnaire was used to assess themaintenanceofparticipantblinding.Themeasureconsistedof 64 statements addressing eight different cognitive per-formance abilities (e.g., recall) asking whether participantsbelieved that they improved, remained the same, or wors-ened specifically because of being in the study. The sumacross all questions was used as a measure of participants’belief in the efficacy of their training arm.Clinical assessors were trained using a standardizedprotocol, and their performance was monitored and cor-rected as necessary throughout the study. A second blindedassessor double-scored assessments.Measures were collected at pre- and posttraining visits.Counterbalanced parallel forms of the RBANS, RAVLT,and RBMTwere used to reduce the potential of test–retesteffects.
RESULTS FROM THE IMPACT STUDY
3 JAGS 2009
Search
Search History:
Searching...
Result 00 of 00
00 results for result for
  • p.
    • Notes
    Load more