FAQ from UK Tour
May 2004AAC Institute Executive Director Katya Hill and board member Barry Romich were in the UK for two weeks in May 2004. (Photos below.) Activities included a number of presentations on "New Tools to Support AAC Evidence-Based Practice". Discussion during these events revealed a few themes that are addressed here as Frequently Asked Questions (FAQs).
Is it necessary to perform a meta-analysis of the research in order to feel that you are providing evidence-based practice?
The three major components of evidence-based practice (EBP) are 1) external evidence; 2) personal evidence; and 3) knowledge and skills of the provider. A meta-analysis is only one method of appraising the external evidence. Several different methods and steps are used in practice to appraise or evaluate research. I’ve attended several conferences on EBP in which faculty have presented various effective and time efficient “tricks of the trade” to perform an appraisal of the research. Performing a meta-analysis, especially when research is limited, may not be the most appropriate use of time and resources for teams. In addition, a publication of a research review or meta-analysis for a group or series of studies may be outdated by the time it goes through the peer-review process or be of limited value in terms of generalizing results to specific clinical questions. Teams have the responsibility to weigh the relevance of the research for each client. Finally, a meta-analysis of research that fails to ask clinical questions will also fail to provide clinical guidance.
To be well-balanced, no one component of EBP should take precedence over another. Synergy is created by effectively applying all three components when making decisions about AAC interventions. Finally, no patient would expect to hear from a health care provider that addressing their problem will have to wait for the research to be performed and analyzed. This does not give teams the excuse to ignore the research; teams must insure that the “best” available evidence is used to support their decisions. Teams starting to apply the principles of EBP shouldn’t feel intimidated by the research component. Start applying the principles using your knowledge and skills and build from there. If you’re monitoring change that results from the intervention, you’ll be tracking the success of EBP also.
Could you elaborate on and/or share a copy of the research review you discussed on recommending an AAC dynamic display to support spontaneous communication?Attached is a handout from my seminar on “Maximizing Performance Using AAC Dynamic Displays: Evidence to Support Practice” that may be used as a reference when a team is considering an AAC dynamic display for an individual. Side one of the handout gives the study, research participants, and major findings for 11 published studies; side two provides the APA references. Click here to download the handout (12K).
Two papers based on research conducted at CATER (Center for Assistive Technology Education and Research) at Edinboro University of Pennsylvania will be presented at the RESNA conference in June 2004. These pilot studies analyzed language samples from individuals who rely on AAC dynamic display systems and report preliminary performance patterns. Results appeared to indicate that using a display scheme based on core and activity rows provided more immediate access to high frequency of occurrence (core) vocabulary stored using semantic compaction. Therefore, individuals who used a fixed-level scheme (semantic compaction) appeared to take better advantage of the immediate access to core vocabulary provided by this display scheme. Individuals with multi-level (page-based) schemes tended to use more multiple key activations to access picture producing vocabulary.
Could you explain selection rate in more detail?Selection rate is one of the 17 quantitative summary measures of the AAC Performance Report. Selection rate is a measure of the ability of the individual to make selections on an AAC system.
Since selection rate directly influences communication rate and communication rate is highly valued by people who use AAC, the clinical value of selection rate is obvious. Selection rate measurement can be used to quantify the performance differences among candidate selection methods. Measuring selection rate using each method and then choosing the method accordingly should result in the highest communication performance. Periodic review of selection rate can point out differences that may relate to positioning, time of day, fatigue, or other factors.
The unit of measure for selection rate is “bits per second”, a term commonly used in the field of human factors and ergonomics (HFE). The size of the array (A) (e.g., number of keys on a keyboard) from which choices are being made determines the number of bits (N) that are available with each choice. A = 2N. N = ln(A) / ln(2). (The symbol ln is the natural logarithm.) The integer number of bits (N) for various array sizes (A) is presented in this chart:
Thus, if a person were able to make one choice per second from a keyboard with 128 keys, the selection rate would be seven bits per second. Selection rate is one of several HFE issues that can have a dramatic impact on communication performance using AAC.
Is the amount of support and learning time between single-meaning pictures and semantic compaction (Minspeak) on AAC systems significantly different?One of the myths of AAC is that one method of representing language is significantly easier to support and learn than another. No one questions that teaching children to read and write requires support and time, and teaching a literate adult to use spelling as an AAC strategy requires minimal support and training. The confusion results from initial impressions of single meaning pictures and semantic compaction. Service providers may believe incorrectly that more support and learning time is needed for AAC systems that are based on the use of multiple meaning icons (semantic compaction).
Let’s start with the external evidence, which is extremely limited. First, no refereed study addresses the question directly. In fact, to my knowledge, no published study has posed support and/or learning time as the leading research questions. So no AAC stakeholder can claim empirically that one method requires more support or learning time over another. Yet, we need to rely on external evidence to help provide insight into this relevant clinical question. What can be implied from the external evidence at various levels of strength to help us answer the question?
Gardner-Bonneau and Schwartz (1989) compared communication speed between traditional orthography (TO) and Words Strategy (WS) with Minspeak. Able-bodied participants were assigned to either the TO group or WS group. Each group was required to perform six traditional typing tests, which measured typing speed. The WS group received 25 hours of training to learn Minspeak before participating in the testing. Results of the experiment showed that by the test 6 the WS rate was over 32 words per minute (wpm), as compared to the TO rate of 25 wpm, which reached asymptote by test 2. While the TO group reached peak performance quickly, the WS group never reached peak as performance improved at every test. The important bit of evidence in this research design is that only 25 hours were required for participants to perform the sentence typing tasks and the learning curve improved with use.
Drager, Light, Speltz, Fallon, and Jeffries (2003) compared 3 system approaches to vocabulary organization: taxonomic grid; schematic grid; and schematic scene on the Dynavox and Freestyle. They found a poor performance across all conditions and a failure to generalize knowledge to learning novel vocabulary for the normally developing children in their study.
In my presentation, you may recall that I cited research and clinical data related to this question. In studying adults considered competent users of AAC, Hill (2001) found that adults using AAC systems with semantic compaction for three months were able to achieve similar performance results to adults using semantic compaction with years of experience. All research participants were self-taught. Also, Hill (2003) documented the vocabulary learning rate of a three-year-old using semantic compaction icons. The child learned 35 icons after 5 hours of intervention. Over the 18 months of intervention, the client learned over 72 icons with a mean length of utterance (MLU) increase from under 1.0 MLU to 3.66 MLU.
An important consideration regarding single meaning picture systems is the amount of support needed for programming vocabulary and messages. While teams may find this approach easy to learn at first encounter, this may not be the case in the long term when one considers time for vocabulary selection, programming, and organization. Activity-based vocabulary approaches can be labor intensive. In addition, ease of use regarding symbol identification may be based on the literacy skills of adults. Children not able to read the text distinguishing like-symbols may not be able to identify vocabulary without prompts, thus failing to achieve independent communication.
EBP means that as health care providers we are careful not to be basing clinical decisions about AAC interventions on impressions of effectiveness, but are searching, appraising, and collecting data to evaluate effectiveness. As you address the question of amount of support and learning time as a factor in recommending an AAC intervention, consider whether you are making a decision based on impressions or an appraisal of data. You should also consider whether you are taking into account the human factor that “ease of use at first encounter frequently doesn’t lead to effective long term use.”
Finally, ask the question: “Isn’t long term effective use more important than ease and speed of learning?”
A warm welcome at Wolfson Centre in London.
Introducing the AAC Institute to the UK audiences.
A beautiful day at the ACE Centre in Oxford.
Deborah Jans and Katya around Storybag library at KeyComm in Edinburgh.
