Response to ASHA's Working Group Report on Auditory Integration Training

In August, 2003 The ASHA Leader published a biased, incomplete and inaccurate report on Auditory Integration Training (AIT) by an ASHA Working Group on AIT. Our response was rejected by The ASHA Leader. Our paper raised serious concerns regarding the accuracy of the Working Group's report, and we feel the report clearly violated ASHA's Code of Ethics (Principle of Ethics: I-G and III-F).

We were not given a specific reason why our response was rejected. ASHA has failed in its duty to provide its members unbiased and accurate information. Since our paper also raised the serious possibility of ethics violations, we were disappointed that the editor of The ASHA Leader did not share our paper with ASHA's Ethics Committee.

We also encourage ASHA members to urge ASHA to recant their policy on AIT, and encourage ASHA to conduct an objective, non-biased review of AIT.

Stephen M. Edelson, Ph.D. and Bernard Rimland, Ph.D.
Autism Research Institute, San Diego, California

ASHA: Retract Your Biased Report on AIT!

Stephen M. Edelson and Bernard Rimland
Autism Research Institute, San Diego, California

Abstract

ASHA's Working Group Report on Auditory Integration Training (AIT) was adopted by ASHA's Executive Board in March, 2003; and was published on the Internet in August, 2003. The Working Group's Report summarized the research findings of only six studies on the efficacy of AIT, although there are 22 other studies on AIT which the Working Group did not discuss. Of the 22 undiscussed studies, 20 had reported positive findings. A careful reading of the Working Group's Report reveals many biases in reporting the results of the six studies— positive findings from three studies supporting AIT were either not reported or were harshly criticized. In contrast, there was no mention of the serious shortcomings of the three AIT studies showing no improvement. ASHA's Working Group Report is in violation of ASHA's Code of Ethics policy which requires accurate, unbiased reporting. Since ASHA's Code of Ethics policy that clearly states that the individual practitioners are responsible for deciding for themselves which interventions are effective, we urge practitioners exercise their right and obligation to continue providing AIT to clients who may benefit. The present authors also urge that ASHA retract the Working Group's Report on AIT and that a scientific and objective review be undertaken.

*****

ASHA: Retract Your Biased Report on AIT!

The ASHA Working Group Report on AIT (Auditory Integration Training), which was published on the Internet in August 2003, fails to provide an objective summary of the available research on AIT. The Working Group erroneously concluded: "Despite approximately one decade of practice in this country, this method has not met scientific standards for efficacy and safety that would justify its inclusion as a mainstream treatment for these disorders" (p. 5). ASHA's Executive Board adopted the Working Group's Report in March 2003. A careful reading of the Working Group's Report reveals many inaccurate and misleading statements. In addition, much important information was excluded. As a result, the ASHA Executive Board, ASHA members, and the general public, were given a biased, inaccurate report on the scientific status of AIT.

The ASHA Report summarized only 6 of 28 research studies. The Report was highly critical of three studies which documented benefits as a result of AIT, but failed to mention the disqualifying shortcomings of the three studies showing no improvement. Following is a brief summary and a discussion of each of the six studies discussed in the ASHA report.

Studies Demonstrating Benefits Conferred by AIT

Rimland and Edelson (1994). This study involved a total of 445 autistic children and adults. The ASHA Report correctly restated the four main questions that the study sought to answer: (a) Does AIT reduce sensitivity to sounds? (b) Is it helpful to use filters (versus no filters) during the AIT listening sessions? (c) Is there a profile that predicts the best candidates for AIT? (d) Are the three different AIT devices equally effective?

One of the important questions addressed in the study was: 'Does AIT reduce sensitivity to sounds?' The Working Group did not discuss these results in their report. Since few autistic children can provide verbal reports, subjects were videotaped while undergoing audiometric testing before and after the AIT sessions. The videotapes were shown to raters who were 'blind' to the 'before/after' conditions. A significant reduction of almost 25% in sound sensitivity was found. The Working Group's failure to discuss one of the most important findings in the study demonstrated their intent to report only results consistent with their own biased view of AIT. Additional examples of this bias are presented below.

Rimland and Edelson (1995). This pilot study involved 17 autistic individuals ranging in age from 4 to 21 years. A significant decrease in behavioral and auditory-related problems was found at the 3-month assessment period for the AIT experimental group (n=9) as compared to the placebo group (n=8). Although the subjects had been matched on age, sex, hearing sensitivity and number of ear infections, there were differences between the two groups that were reflected in the baseline scores. Two different methods were used to correct statistically for this a priori discrepancy.

The ASHA Report was highly critical of the first statistical analysis, which used difference scores to compensate for the discrepancy in baseline scores, and described the analysis as "questionable." The ASHA report failed to mention that a second method, a statistical slope analysis, had also been used to help correct for the baseline differences (Rimland and Edelson, 1995, p. 67). This supplementary analysis was recommended by statisticians at Stanford University and MIT, who were credited for their help on the first page of the report. The slope analysis, which was not mentioned by the ASHA Working Group, confirmed the conclusions of the difference score analysis.

Edelson et al. (1999). Nineteen autistic individuals participated in this study, nine in the AIT experimental group and ten in the placebo (control) group. Behavioral measures indicated a significant decrease in behavior problems at three months for the experimental group compared to the control group.

Additionally, the study documented a dramatic improvement (i.e., normalization) in brain wave activity (P300) in response to an auditory stimulus in three subjects who received AIT. There were no improvements in brain wave activity in two subjects who received the placebo treatment. The ASHA report criticized this important finding because of the small number of subjects tested. P300 measures are very difficult to obtain from autistic individuals. Published reports using P300 technology often include only one or two subjects. The small number of subjects evaluated in this study is a respectable sample size, especially since P300 is an objective measure of brain wave activity. The Working Group's failure to acknowledge the significance of this finding is but another example of the Group's anti-AIT bias.

Studies Demonstrating No Improvement as a Result of AIT

Although the ASHA Report mentioned a few of the shortcomings of the "no improvement" studies, they did not discuss the very serious problems which invalidate their conclusions. The ASHA committee should have been aware of the problems discussed below because they were explicitly discussed in the Edelson et al. (1999, pp. 74-75) study, a paper cited and discussed in the ASHA report.

Bettison (1996). In this study 40 autistic children participated in the experimental (AIT) and 40 in the control (placebo) group. Both groups were said to have improved over a 12-month period. The author suggested that the improvement in both groups may be due to listening to music in a structured environment. The ASHA Report raised the possibility that '... an effect other than a specific treatment (i.e., AIT) was responsible for the improvements" (p. 4).

The ASHA Report failed to mention that the measures used in Bettison's study were clearly inappropriate, as had been pointed out by Edelson's critique of her study (Edelson et al., 1999). Sound sensitivity was measured using a simple questionnaire designed by Rimland and Edelson (1991) to explore the phenomenon of hearing sensitivity. The questionnaire was not designed to measure treatment effectiveness. The scoring method devised by Bettison, not by Rimland and Edelson, lacks even face validity. For example, if a parent agreed with the statement: "Have there been certain sounds which the person does not seem to hear?," this response was scored as hypersensitivity to sounds rather than hyposensitivity to sounds.

The other instrument used in the study was the Developmental Behavior Checklist, which was, like the Hearing Sensitivity Survey, not designed to measure treatment effectiveness. Given the serious problems with these assessment measures, one cannot conclude, in any way, that AIT was effective nor ineffective. In other words, neither of the assessment measures were appropriate measures of differences between the AIT and placebo groups. This disqualifying problem was not mentioned in the ASHA report.

Zollweg, Palm, and Vance (1997). This study involved 15 subjects who participated in the experimental (AIT) group and 15 subjects who participated in the control (placebo) group. No significant differences between the two groups were reported.

Most of the subjects in this study were mildly to profoundly retarded, and fewer than one- third were autistic (Gelman, 1996). No one has claimed that AIT is a treatment for mental retardation.

When describing the Zollweg et. al (1997) study, the authors of the ASHA Report showed that they were aware of the candidacy problem in this study by writing: "Although AIT is typically considered to be a treatment for individuals with autism spectrum disorders, such treatment has been recommended for other disorders (e.g., learning disabilities, attention deficit disorders, depression) (ASHA, 1994; Berard, 1993)" (p. 4). None of the subjects in the Zollweg et al. (1997) study were diagnosed as learning disabled, attention deficit disordered, or depressed. The ASHA Report failed to mention that the majority of subjects were mentally retarded, and less than one-third were autistic.

Additionally, the volume level of the AIT listening sessions was much higher than recommended. The recommended level is 80 dB SPL or lower, whereas the volume level in the Zollweg et al. (1997) study was as high as 122 dB SPL (p. 43). Even if AIT could help mentally retarded individuals, the extremely loud sound level would be expected to preclude the possibility of any positive effects from the listening sessions. The ASHA Report overlooked these problems.

Yencer (1998). This study was based on 36 children with Central Auditory Processing Disorder (CAPD). Half were assigned to an experimental (AIT) group and the other half was assigned to a control (placebo) group. No differences were found between the two groups one month following the AIT listening sessions on a number of different brainwave, audiological and behavioral measures.

The Yencer study assessed differences for only one month. It is well established that the improvements conferred by AIT take up to three months to fully appear (Berard, 1993; Edelson et al., 1999; Rimland & Edelson, 1995). None of the studies demonstrating positive results discussed above found a difference at one-month following the listening sessions. Furthermore, a study by Huskey et al. (1994), using a similar subject population, CAPD, which was not mentioned in the ASHA Report, did not find any improvement 4 to 6 weeks following AIT, but did report significant improvements 6 to 8 weeks following AIT.

The ASHA Report failed to mention the inappropriateness of assessing changes at only one month following AIT, even though the Report cited several papers that emphasized that changes typically occur up to three months following AIT (Berard, 1993, p. 91; Rimland and Edelson, 1995, p. 62; Edelson et al., 1999, pp. 74 & 75). Yencer had been advised by one of us (SME) that a three-month follow-up testing was required; she nevertheless chose a one- month follow-up for her personal convenience.

The ASHA committee was responsible for being aware of the major problems in the three studies cited above. Their failure to mention these problems in their report is inexcusable.

Twenty Uncited Studies Support the Efficacy of AIT

The ASHA report cited only six research studies, even though there were many more studies supporting the use of AIT, some of which had been published in ASHA journals, ASHA sponsored conferences, and other professional conferences and newsletters. Edelson and Rimland recently completed a review of all 28 known studies on AIT (Edelson & Rimland, Submitted for publication). They found 23 studies (83%) supporting the efficacy of AIT, three studies (11%) claiming no evidence of efficacy, and two studies (7%) reporting ambiguous, contradictory results. Read the AIT review paper.

Treatment Risks

The ASHA report described in great detail the findings by Rankovic, Rabinowitz, and Lof (1996) in which they measured sound levels as high as 118 dB SPL in one AIT device and suggested that AIT may be detrimental to hearing. As stated earlier, the recommended sound level is 80 dB SPL or lower. Many electronic devices available to the public, such as radios and audiocassette players, also have the potential for very high output levels – should their use be banned? It is the responsibility of the practitioner to ensure that the sound level during AIT is played at a safe and appropriate level. Rankovic et al (1996) based their findings on the practice of a single AIT practitioner. It is unreasonable for Rankovic et al. and the ASHA Working Group to condemn all AIT practitioners.

Ethical Issues

ASHA's Working Group Report is in violation of ASHA's 'Code of Ethics' (January 1, 2003). It is very clear that the Working Group's Report misrepresented the research findings on AIT. Principle of Ethics III-F states that: "Individuals' statements to the public—advertising, announcing, and marketing their professional services, reporting research results, and promoting products—shall adhere to prevailing professional standards and shall not contain misrepresentations."

Note that Principle of Ethics I-G states: "Individuals [italics added] shall evaluate the effectiveness of services rendered and of products dispensed and shall provide services or dispense products only when benefit can reasonably be expected." Thus, ASHA places the responsibility for evaluating services directly on the providers of the services, who are required to render their own judgment as to the value of those services.

Although we believe the report of the ASHA Working Group to be so fatally flawed that it should be retracted, we are pleased that the ASHA 'Code of Ethics' not only permits, but in fact requires, that ASHA members determine for themselves whether or not to provide AIT services.

As stated in the ASHA report, the majority of AIT practitioners are speech-language pathologists and audiologists (p. 1). Unless it is rescinded -- and it should be -- ASHA's policy regarding AIT may have a serious negative effect on the professional services provided by ASHA members and the well-being of many autistic children and adults.

Those who respect objective research, and who are dedicated to helping their clients, will no doubt choose to continue the practice of AIT.

Conclusions

The ASHA Report, based on a selective misreading of six studies, is biased, incomplete, and highly inaccurate. We urge that it be retracted and that a new, more comprehensive and objective review be undertaken. In the meantime, since there is abundant evidence from our review of 28 studies that AIT is beneficial to many autistic children and adults, we urge that AIT practitioners continue to offer their invaluable AIT services. To discontinue the practice of AIT would clearly be inimical to the autistic individuals and their families.

References

ASHA (2004). Auditory integration training. ASHA Supplement 24, in press.

ASHA. (2003, Aug. 5). ASHA adopts AIT policy. THE ASHA Leader, 8, 3.

Berard, G. (1993). Hearing equals behavior (Trans.). New Caanan, CT: Keats. Original work published in 1982).

Bettison, S. (1996). The long-term effects of auditory training on children with autism. Journal of Autism and Developmental Disorders, 26, 361-374.

Edelson, S.M., & Rimland, B. A Review of 28 Studies on Auditory Integration Training: Ample Evidence Supports Its Efficacy. San Diego: Autism Research Institute.

Edelson, S.M., Arin, D., Bauman, M., Lukas, S.E., Rudy, J.H., Sholar, M., & Rimland B. (1999). Auditory integration training: a double-blind study of behavioral, electrophysiological, and audiometric effects in autistic subjects. Focus on Autism and Other Developmental Disabilities, 14, 73-81.

Gelman, J. (1996, January 8). Reining in AIT: Clinicians call for stricter guidelines, more research. ADVANCE for Speech-Language Pathologists & Audiologists, pp. 4 & 11.

Huskey, B., Barnett, B., & Cimorelli, J.M. (1994). The effects of auditory integration therapy on central auditory processing. Paper presented at the American Speech Language-Hearing Conference, New Orleans.

Rankovic, C.M., Rabinowitz, W.M., & Lof, G.L. (1996). Maximum output intensity of the Audiokinetron. American Journal of Speech-Language Pathology, 5, 68-72.

Rimland, B., & Edelson, S.M. (1991). Hearing sensitivity questionnaire. San Diego: Autism Research Institute.

Rimland, B., & Edelson, S.M. (1994). The effects of auditory integration training in autism. American Journal of Speech-Language Pathology, 5, 16-24.

Rimland, B., & Edelson, S.M. (1995). Brief report: A pilot study of auditory integration training in autism. Journal of Autism and Developmental Disabilities, 25, 61-70.

Yencer, K.A. (1998). The effects of auditory integration training for children with Central Auditory Processing Disorder (CAPD). American Journal of Audiology, 7, 32-44.