iPad Based Software Platform for Language amp Cognitive Rehabilitation Swathi Kiran Carrie Des Roches Isabel Balachandran Stephanie Keffer Elsa Ascenso Anna Kasdan ID: 211283
Download Presentation The PPT/PDF document "Validating Patient Outcomes Using an" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Validating Patient Outcomes Using an iPad-Based Software Platform for Language & Cognitive Rehabilitation
Swathi
Kiran, Carrie Des Roches, Isabel Balachandran, *Stephanie Keffer, Elsa Ascenso, *Anna KasdanSpeech and Hearing Sciences, Boston UniversityDepartment of Neurology, Massachusetts General Hospital
Funding from Wallace H. Coulter Foundation: BU-Coulter Translational Partnership Program
1
ASHA 2013Slide2
Disclosure-Swathi Kiran
Has significant financial InterestChief Scientist for Constant TherapyOwnership stock in Constant Therapy2ASHA 2013Slide3
Other Authors: Carrie Des Roches, Isabel Balachandran
, Elsa AscensoNothing to discloseSignificant contributorsStephanie Keffer, Anna KasdanNothing to discloseASHA 2013
3DisclosureSlide4
IntroductionAbout 795,000 Americans each year suffer a new or recurrent stroke (NIDCD.gov). Also, about 1.7 million individuals suffer from traumatic brain injury each year (CDC.gov).
Individuals with language and cognitive deficits following brain damage likely require long-term rehabilitation. Consequently, it is a huge practical problem to provide the continued communication therapy that these individuals require. 4ASHA 2013Slide5
Using technology to improve treatment delivery
Recent studies have examined the efficacy of rehabilitation techniques, such as videoconferencing, for individuals with hearing, stuttering and motor speech issuesOther studies have provided aphasia therapy over the internet to individual patientsMore recently, there have several computerized brain-training software designed for normal adults. 5
ASHA 2013Slide6
What is the evidence behind using technology to deliver treatment?
CogMedA software targeted at improving working memory abilities in individuals with brain injury (Johansson & Tornmalm M, 2012; Lundqvist et al. 2012). These studies found improvements in working memory skills on the trained CogMed software as well as on other working memory tasks and functional settings. . Posit Science
Barnes et al (2009) examined the effectiveness of the software Posit Science in improving auditory processing speed in individuals with mild cognitive impairment (MCI). Although differences between the experimental and control group were not statistically significant, verbal learning and memory measures were higher in the experimental group than the control group. 6ASHA 2013Slide7
What is the evidence behind using technology to deliver treatment? Lumosity
Finn and McDonald (2011) used Lumosity software to target attention, processing speed, visual memory in experimental and waitlisted controls. Results showed experimental participants improved on the training exercises more than the controls. There are other software programs- that function more like AAC- devices.Therefore, there an increased awareness and momentum for applying computer technology in the rehabilitation of aphasia7
ASHA 2013Slide8
Rationale
Additionally, there is increased patient demand to transition from traditional but outdated flashcard based therapy in order to keep up with the evolution of technology. Nonetheless, the burden of evidence for technology-based treatment applications is no different than traditional treatment approach for rehabilitation after brain damageQ1. Can we provide a technologically based rehabilitation program that meets the same benchmarks for clinical efficacy?Q2. How do we individualize treatment for patients with brain damage as no two patients are alike?8ASHA 2013Slide9
Study
Question: Does a structured therapy program that includes homework practice delivered through an IPAD result in significant gains in overall communication? Goal: Compare patients who receive a structured IPAD delivered therapy program that is practiced up to 7 days a week with patients who receive standard one-on-one individualized therapy that is provided 1 or 2 days per week by a therapist. 9
ASHA 2013Slide10
StudyBecause of the flexibility that
ipads provide to patients and the accessibility to free/paid apps that provide variable levels of exercises, it is important to standardize the nature and form of treatment that is provided to patients using ipads. Since patients have access to ipads at home, it provides a unique opportunity to examine the extent of compliance when patients are provided with a homework regimen10ASHA 2013Slide11
Participants
Experiment (N = 40)Control (N= 9)Ave Age63 years (SD = 11)
68 years (SD = 10)Ave Months post Onset 54 months (SD = 47)98 months (SD = 132)WAB – AQ (western aphasia battery)68.5 (SD = 26.52)68.1 (SD = 31.35)CLQT Composite Severity (cognitive linguistic quick test)64% (SD = 25)54% (SD = 28)11ASHA 2013Slide12
Demographic data
12ASHA 2013Slide13
ASHA 2013
13Slide14
ASHA 2013
14Slide15
Experimental Design
Pre-Tx Assessment
Pre-Tx AssessmentPost -TxAssessment
Weeks
Post -
T
x
Assessment
EXPERIMENTAL PATIENTS (N = 40)
CONTROL PATIENTS (N = 10)
15
ASHA 2013Slide16
16
ASHA 2013Slide17
17
ASHA 2013Slide18
Sample therapies/assessments for language and cognitive processing
ASHA 2013Slide19
Structure of the tasks- Language
19ASHA 2013Slide20
Structure of the tasks- Cognitive
20ASHA 2013Slide21
21
ASHA 2013Slide22
Individualized therapy assignment based on initial performance
22ASHA 2013Slide23
Reading Passages (Level 3)
Picture Ordering Tasks (5 items)Map Tasks (10 items, Level 2)
Syllable Identification (10 items)Picture Spelling (Level 3)Word IdentificationWord Copy (10 items)Category MatchingSound Identification (10 items)Clock Tasks (10 items)Week1Week2
Week3
Week4
Week5
Week6
Week
7
Week
8
Week
9
Week
10
Picture Spelling (Level 2)
Addition (Level 3)
Addition (Level 4)
Addition (Level 4)
Addition (Level 4)
Addition (Level 5)
Syllable Identification
Syllable Identification
Syllable Identification
Multiplication (Level 4)
Reading Passage (Level 2)
Picture Spelling (Level 3)
Picture Spelling (Level 3)
Picture Spelling (Level 3)
Picture Spelling (Level 4)
Picture Spelling (Level 5)
Multiplication (Level 2)
Multiplication (Level 3)
Multiplication (Level 3)
Picture Ordering Tasks (Level 4, 10 items)
Word Ordering (Level 2)
Subtraction (Level 3, 5 items)
Subtraction (Level 3)
Subtraction (Level 3)
Subtraction (Level 4)
Subtraction (Level 5)
Picture Ordering Tasks
Picture Ordering Tasks (Level 2)
Picture Ordering Tasks (Level 3)
Division (Level 3)
Reading Passage (Level 3)
Reading Passage (Level 3)
Reading Passage (Level 3)
Syllable Identification
Syllable Identification
Division
Division (Level 2)
Clock Math (Level 2)
Word Ordering (Level 3)
Word Ordering (Level 3)
Word Ordering (Level 3)
Multiplication (10 items)
Multiplication (Level 2, 5 items)
Word Matching (10 items)
Map Tasks (Level 2)
Map Tasks (Level 3)
Map Tasks (Level 3)
Word Ordering Task (Level 4)
Word Ordering Task (Level 5)
23
ASHA 2013
#
29 during week 6 homework
# 25
during week
4
homework
# 44 for all 10 weeksSlide24
Based on WAB, CLQT
24ASHA 2013Slide25
Low Language profile- Low cognitive profile
High Language profile- low cognitive profileLow language profile- high cognitive profile
High language profile- high cognitive profile86 year old male75 year old male77 year old male56 year old malecategory matchingfeature matchingpicture naming rhymingsound identificationword identificationsound to letter matching word copy picture matchingsymbol cancellationpicture spellingnaming picture
clock reading
instruction sequencing
picture ordering
sound matching
symbol matching
voicemail
category identification
category matching
feature matching
letter to sound matching
reading passage
sound identification
sound to letter matching
word copy
word spelling
word ordering
category matching
feature matching
letter to sound matching
sound to letter matching
map reading
picture spelling
reading passage
rhyming
Sound
identification
syllable identification
word spelling
word
problems
25
ASHA 2013Slide26
26
ASHA 2013Slide27
Carrie/Please insert snapshots of the patient dashboard- that shows start therapy, we will now do…
27
ASHA 2013Slide28
28
ASHA 2013Slide29
MethodsDuring the weekly clinic sessions, the clinician would decide to continue the participant on the same task or to modify the treatment plan based on his/her performance
. If the participant achieved 95% or higher accuracy two times in succession,The clinician would either progress the next level of difficulty (e.g., Addition Level 1 to Addition Level 2) Would progress to a different task (e.g., assign category identification after category matching).If participants performed at low accuracies or no change over several sessions, that therapy task was replaced with another task from the task list.
29ASHA 2013Slide30
30
ASHA 2013Slide31
Overview of data analysis
Total therapy duration in weeksCompliance- weekly log in timesIndividual patient level analysisAnalyze by weekAnalyze by taskAnalyze by itemOverall patient performance over timePatient performance over time relative to population mean
Group level analysis: Analysis of tasks by patientsGroup level analysis: Analysis of task by items, co-factorsChanges on standardized measuresASHA 201331Slide32
The average therapy period for controls was 12.964 weeks Average therapy period for experimental patients was 12.567 weeks
32ASHA 2013
1. Total therapy duration in weeksSlide33
2. Compliance- Rates of log in to therapy
33
Control patients
Experimental patients
ASHA 2013Slide34
Patient usage by week
Legend:
Control patientsExperimental patients34ASHA 2013Slide35
3. Individual patient level analysis: By week
35
ASHA 2013Slide36
ASHA 2013
363. Individual patient level analysis: By taskSlide37
37
ASHA 2013
3. Individual patient level analysis: By itemsSlide38
38
ASHA 2013
4. Historical individual performance Slide39
Time
39
ASHA 2013
4. Historical individual performance Slide40
5. Patient performance over time relative to population mean
40ASHA 2013Slide41
41
ASHA 2013Slide42
42
ASHA 2013Slide43
43
ASHA 2013Slide44
44
ASHA 2013Slide45
6. Group level analysis: Analysis of tasks by patients- Individual
Analysis Quantifying ChangeSlide46
Individual analysis: Quantifying ChangeWith tasks that had an R
2 value of above 0.25, the average of the first two sessions was subtracted from the average of the last two sessionsThen each of those values was determined to be a “good” or a “bad” changeA “good” change in accuracy was any value above 0A “good” change in latency was any value below 0Slide47
Individual subject analysis by task
Low Language profile- Low cognitive
profileHigh Language profile- low cognitive profileLow language profile- high cognitive profileHigh language profile- high cognitive profile47ASHA 2013Slide48
Group analysis of task improvement
Average % Significant (Overall)19.63%
% of 19.63% considered to be "Good"79.62%48ASHA 2013Slide49
Change in GOOD Significant R2
Accuracy (%)
Average14.45%Smallest Change0.48%Largest Change74.86%Group analysis of task improvementSlide50
Change in GOOD Significant R2
Latency (seconds)
Average0.144538506Largest Change0.74861Smallest Change0.004765Group analysis of task improvementSlide51Slide52
7. Group level analysis- co-factors
Mixed regression models for each treatment taskFor e.g., for word identificationOverall effect of treatment is significant
52ASHA 2013Estimates for Accuracy LabelEstimateSEDFt valuePr > |t|Conditional improvement
0.01961
0.0091
294
2.15
0.0325
Effect of WABAQ on improvement
-0.00066
0.0001
294
-4.88
<.0001
Effect of Composite Severity on improvement
0.01820
0.0137
294
1.32
0.1874
Task
Accuracy
Latency
Estimate
p
value
Estimate
P value
Category Identification
-0.00633
0.5397
0.1135
0.569
Category Matching
-0.00019
0.9641
-0.05835
0.2064
Feature Matching
0.02356
0.0034
-0.1878
0.0109
Letter to sound matching Level 2
0.01018
0.294
-0.5329
0.0002
Rhyming
0.01521
0.0032
-0.0223
0.6982
Word spelling completion LV 1
0.01456
0.0187
-0.09176
0.3882
Word Identification
0.01961
0.0325
0.1173
0.0183
Addition Level 1
0.122
0.3455
-5.9979
0.0847
Addition LV2
-0.05748
0.0307
2.8665
0.0093
Addition LV3
-0.1141
0.1456
1.6905
0.1655
Addition LV4
-0.1361
0.1438
-3.1877
0.0719
Addition LV5
0.02881
0.1826
0.1198
0.8927
Clock Reading Level 1
0.000816
0.9405
-0.4324
0.0016
Clock Math
Lv
2
-0.03675
0.9272
33.3335
0.0458
Clock Math LV3
0.01032
0.9718
Division LV1
0.05175
0.6368
-0.7804
0.7296
Division LV4
0.48
0.9587
618.69
0.3172
Instruction sequencing
0.03433
0.6964
4.6205
0.0971
Map Reading LV1
0.2592
0.0014
0.09436
0.9073
Map Reading LV2
-0.1268
0.1432
16.2842
<.0001Slide53
7. Can treatment outcome be predicted for each patient?
Solution for Random Effects
PatientIdEstimateStd Err PredDFt ValuePr > |t|23-0.072160.2081589-0.350.7289
24
-0.5612
0.1757
589
-3.19
0.0015
903
0.4057
0.2544
589
1.59
0.1113
913
0.3272
0.3258
589
1.00
0.3157
955
0.08395
0.2471
589
0.34
0.7342
978
1.0518
0.2442
589
4.31
<.0001
1049
-0.9511
0.1953
589
-4.87
<.0001
1079
-0.02336
0.2463
589
-0.09
0.9245
1091
-0.3460
0.2881
589
-1.20
0.2303
1339
0.4617
0.2044
589
2.26
0.0243
This patient show significantly less improvement than the average improvement
This patient show significantly more improvement than the average improvement
53
ASHA 2013Slide54
8. Mean Changes on standardized test performance
54ASHA 2013Slide55
8. Change on standardized tests
SubtestExperimental Group (N= 40)Control Group (N = 9)WAB-LQ
2.13% (t = -2.05, p <.05)1.42% (t = 1.07, ns)WAB-CQ2.60% (t = -2.05, p <.05)1.32% (t = 1.03, ns)WAB-AQ4.14% ( t = 3.11, p <.01)0.65% (t = .76, ns)CLQT-composite severity9.12% (t = 3.28, p < .01)4.44 % (t = .76, ns)CLQT-Attention11.3 % (t = -.4.55, p <.0001)7.6% (t = -1.19, ns)CLQT-Memory1.55% (t = -0.84, ns)1.140 (t = 0.30, ns)CLQT-Executive Function6.41% (t = 3.15, p < .01)1.66 (t = 0.52, ns)CLQT- Language
1.42% (t = 1.15, ns)1.65 (t = 1.26, ns)CLQT-
Visuospatial skills7.81 (t = 3.43, p < .001)
2.96 (t = 0.61, ns)
55
ASHA 2013Slide56
Summary of results
Experimental and control patients completed 11 weeks of treatmentPatient compliance was high-ranged from 1 time/week to 11 times/weekPatients who logged in more often showed more changes on tasks assignedIndividual patient analysis by items, task, weekly- gives insight into how patients performChanges in accuracy and latency are across tasks Changes seen on standardized tests for experimental patients, less for control patientsASHA 2013
56Slide57
Thank you ! Questions?
ASHA 201357Slide58
ReferencesBarnes D,
Yaffe K, Belfor N, Jagust W, DeCarli C, Reed B, Kramer J. Computer-Based Cognitive Training for Mild Cognitive Impairment: Results from a Pilot Randomized, Controlled Trial. Alzheimer Disease and Associated Disorders 2009; 23(3), 205-210.Finn M, McDonald S. Computerised cognitive training for older persons with mild cognitive impairment: A pilot study using a randomised controlled trial design. Brain Impairment 2011; 12(3), 187–199.Johansson B, Tornmalm M. Working
memory training for patients with acquired brain injury: Effects in daily life. Scandanavian Journal of Occupational Therapy 2012; 19(2), 176-83. doi:10.3109/11038128.2011.603352.Lundqvist A, Grundström K, Samuelsson K, Rönnberg J Computerized training of working memory in a group of patients suffering from acquired brain injury. Brain Injury 2010; 24, 1173-1183. doi:10.3109/02699052.2010.498007ASHA 201358