RESNA 26th International Annual Confence

Technology & Disability: Research, Design, Practice & Policy

June 19 to June 23, 2003
Atlanta, Georgia


AAC PERFORMANCE DATA FOR AN INDIVIDUAL WITH AMYOTROPHIC LATERAL SCLEROSIS

Stephanie Cook and Katya Hill
Edinboro University of Pennsylvania
Edinboro, Pennsylvania 

ABSTRACT

A central component of evidence-based practice (EBP) is the collection of quantitative data at the personal level. Persons with amyotrophic lateral sclerosis (ALS) may have rapidly changing needs regarding deterioration of physical access to an augmentative and alternative communication (AAC) system. Methods and tools are available to measure selection and communication rates to compare access methods. The collection and analysis of performance data provides clinicians with the quantitative data needed for EBP. This case study exemplifies the use of performance data for making clinical decisions regarding alternative access to an AAC system.

BACKGROUND

According to the Amyotrophic Lateral Sclerosis Association (ALSA) (2001), an estimated 30,000 Americans have ALS at any given point in time. Due to the deterioration of motor skills over the course of the disease, many of these people will experience the loss of all natural speech. Although individuals with ALS who rely on AAC devices are provided with speech output assistive technology, their communication performance is typically far lower than individuals with normal speech (1).

Currently, speech-language pathologists are expected to provide clinical services in accordance with the principles of evidence-based practice (EBP). EBP requires the use of both external (field) evidence and evidence at the personal level. In order to obtain "patient oriented evidence that matters"(POEMS), quantitative data is required (2). Quantitative data related to the communication competence of an individual who relies on AAC, requires performance measurement based on language sampling. Three summary measures that relate to an individual's ability to access an AAC device include: selection rate, communication rate, and frequency of errors.

For many clinicians, decisions pertaining to comparisons among access methods have been based on clinical intuition, trial-and-error counts, or have not been documented at all. Tools and methods to document selection rate (3) and communication rate (4) are available for clinical use.

Professionals working with persons with ALS often make decisions about the benefits of available strategies without the support from a quantitative research base. The current availability of standardized methods for measuring and calculating specific summary measures could provide valuable assistance in the areas of clinical intervention, outcomes measurement, and research.

RESEARCH QUESTION

Since 1999, language activity monitoring has been available for AAC systems to support the collection and analysis of language samples. However, limited, if any, performance data based on logfiles are available to support evidence-based practice and guide clinical decision-making. The purpose of this single subject case study was to determine whether performance data for the dependent variables provided information to customize the selection technique to optimize communication performance.

METHOD

This single subject case study is an A-B-C... design with changing conditions (Richards, et.al 1999). The subject is a 70 year-old white male with amyotrophic lateral sclerosis. The subject presented himself during the initial assessment for an AAC system in Stage 5 (no useful speech). The subject had no residual speech or writing for communication and was using a low-tech alphabet board for spelling.

Performance data were collected for the subject's use of a 128-location, synthesized speech hybrid AAC system (Pathfinder? with AQLS?) using the LAM and observational recording of responses. LAM logfiles were uploaded into the computer for analysis and reporting using PeRT (Performance Report Tool©). Inter-judge reliability of the transcription process was 100% agreement for the entire set of logfiles.

The results in this paper are limited to documenting the performance related to selection rate (bits per second), communication rate by language representation method (words per minute), and frequency of errors for spelling as dependent variables. Language sampling from interactive communication in the daily environment was used to measure the variables. Performance is reported on the baseline data (A) for direct keyboard selection. Data for the changing conditions (B, C, D, E, F) show performance using 1-switch scanning and modifications to maintain access to the AAC system. Data reflect one-week intervals required for clinical services with the introduction of intervention B, or 1-switch scanning.

RESULTS

A two-hour training session was necessary for subject to learn basic operational skills of the language software and device. At week 6, subject could no longer direct select on the keyboard. Customization of the Pathfinder? continued to be tailored to the subject's individual needs over the duration of the remaining six-week period to maintain optimal performance. During this time, feature customization was needed to adjust the settings to meet the subject's changing abilities.

Table 1 shows the results of the dependent variables for the baseline performance (direct keyboard selection) and subsequent performance for 1-switch scanning. No significant difference exists in selection rate among the sampling contexts. The subject was able to maintain access to the Pathfinder? for communication through week 13 with customization to the available scanning options.

Table 1. Results of performance for changing conditions required for 1-switch scanning.

 

Conditions

 

A1

A2

B

C

D

E

F1

F2

 

Keyboard

1-switch scanning

Performance Data

Week 1

Week 6

Week 8

Week 9

Week 10

Week 11

Week 12

Week 13

Selection Rate(bits/sec)

2.19

.00

0.65

0.62

0.60

0.68

0.63

0.57

Communication Rate by Spelling
(words/min.)

1.96

0

0.21

0.65

0.28

0.60

0.70

0.62

Errors of Spelled Words (%)

24

0

12

37

15

10

53

10

Table 2 depicts the features that were customized, representing the changing conditions, to maintain the alternative access method to the Pathfinder using 1-switch scanning. The selection technique menu options that remained at the default setting throughout these 6 weeks were: 1) row/column scan direction; 2) left-right scan direction; 3) predictive selection; 4) acceptance time; and 5) release time.

Table 2. Scanning Menu features customized on the Pathfinder

 

Changing Conditions for 1-Switch Scanning

 

B

C

D

E

F

F

Parameter

Week 8

Week 9

Week 10

Week 11

Week 12

Week 13

Scanning Delay

2.0

1.5

1.5

1.0

1.0

1.0

Auto Repeat

Off

Off

Off

Off

On

On

Number of Rescans

3

2

2

1

1

1

On-Screen Scanning

Last Position

Last Position

First Position

First Position

First Position

First Position

The scanning delay of the device was reduced to allow a faster rate as the subject became more familiar with AQLS? which changed for Condition C and E. In addition, the number of rescans was reduced for week 9 (Condition C), and again at week 11 (Condition E) as subject experienced increased familiarity. The Auto Repeat option, which allows a subject to repeat a key selection, was activated at week 12, or Condition F. The on-screen scanning configuration began initially in last position, but moved to first position as the activity row was utilized more frequently by the subject for Condition D. Overall the frequency of errors for spelling was variable.

DISCUSSION

The data for this study document performance measures for changing access to an AAC system for an individual with ALS. Due to the rapid decline of the subject's condition, frequent modifications were made to further customize the device to his ongoing needs. The subject's strengths and weaknesses were assessed through quantitative data analysis allowing for modifications as needed. With the use of evidence-based practice to customize AAC devices for persons with ALS, there can be a greater communication potential for this population.

REFERENCES

  1. Hill, K., Romich, B., & Cook, S. (2002). Asking the right questions for evidence-based practice and people with ALS. In Proceedings of the CSUN Conference, Los Angeles, CA.
  2. Dollaghan, C. (2002). An evidence-based approach to clinical practice in communication disorders. Presented at American Speech-Language-Hearing Association Convention. Atlanta, Georgia.
  • Romich, B.A., Hill, K.J. & Spaeth, D.M. (2001). AAC selection rate measurement: A method for clinical use based on spelling. In Proceedings of the 2001 RESNA Annual Conference. Arlington, VA: RESNA Press. 52-54.
  • Romich, B.A., & Hill, K.J. (2000). AAC communication rate measurement: Tools and methods for clinical use. In Proceedings of the 2000 RESNA Annual Conferrence. Arlington, VA: RESNA Press. 58-60.
  • Richards, S.B., Taylor, R.L., Ramasamy, R., Richards, R.Y. (1999). Single subject research: Applications in educational and clinical settings. San Diego: Singular Publishing Group, Inc.
  • CONTACT INFORMATION

    Stephanie Cook
    CATER 102 Compton Hall
    Edinboro University of Pennsylvania
    Edinboro, PA 16444

    E-mail: Cassieson1@aol.com

    RESNA Conference Logo