4 Separate Papers. Write a one-page review, single spaced of each of the assigned readings. Article responses should include a description of the article itself and a reaction to the article. I will a

However, because of the diverse in¯uences on parents' decisions to obtain the interventions, a multifaceted strategy may be needed. Copyright# 2000 John Wiley & Sons, Ltd. Applied behavior analytic (ABA) treatment for children with autism has been the topic of many hundreds of studies since the early 1960s (Matson, Benavidez, Compton, Paclawskyj, & Baglio, 1996) and has long been recognized as the intervention having the most empirical support for these children (e.g., DeMyer, Hingtgen, & Jackson, 1981). Nevertheless, children with autism who participate in ABA treatment may also receive various other forms of intervention, such as speech therapy (Prizant, Schuler, Wetherby, & Rydell, 1997), Sensorimotor Integration Therapy (Ayres, 1972), play therapy (Greenspan, 1992), psychotropic medication (Campbell & Hueva, 1995), and alternative medicine, including elimination diets and megavitamin therapies (Rimland & Baker, 1996). These interventions may complement behavior analytic treatment by enhancing children's skills or reducing maladaptive behaviors that hinder children's progress, such as aggression or stereotypies. Conversely, they may undermine Copyright# 2000 John Wiley & Sons, Ltd.

Behavioral Interventions Behav. Intervent.15: 83±97 (2000) * Correspondence to: Tristram Smith, Department of Psychology, Washington State University, P.O. Box 644820, Pullman, WA 99164-4820, USA. E-mail address: [email protected] this treatment by diverting limited resources into interventions that are ine€ective or even harmful (Jacobson, Mulick, & Schwartz, 1995).

Given the potential costs and bene®ts of these supplemental interventions, information about their use is important to obtain. Though investigators have noted that supplemental interventions appear quite common (e.g., Green, 1996), precise data are sparse with regard to the rate at which children with autism receive these interventions, when and why their parents select them, and how e€ective parents perceive them to be. Aman, Van Bourgondien, Wolford, and Saphare (1995) found that, in a community sample, individuals with autism were more likely to receive standard psychotropic medications (such as the ones mentioned above) than alternative medicine. However, this ®nding may not apply to individuals in ABA treatment because the combination of such treat- ment with alternative medicine is often recommended to families (e.g., Rimland, 1998). Informal surveys have suggested that families may evaluate alternative medicine interventions such as megavitamins more favorably than standard psychotropic medicine (e.g., Rimland, 1987), but it is unclear whether this ®nding would be replicable in a more comprehensive investigation. A peer-reviewed survey indicated that parents gave positive ratings to one sensorimotor therapy, Auditory Integration Training (AIT), which is intended to reduce hypersensitivity to sounds (Rimland & Edelson, 1987). However, well studied ABA procedures also exist to address this problem (Charlop-Christy & Kelso, 1997, pp. 169±175).

Therefore, parents of children in ABA treatment may have a di€erent perspective on AIT than the parents in the study by Rimland and Edelson (1994).

The purpose of the present investigation was to obtain additional infor- mation on supplemental interventions received by children who were par- ticipating in one form of ABA treatment (the UCLA Treatment Model; Smith, Donahoe, & Davis, in press). Study 1 surveyed a wide range of supplemental interventions, while study 2 focused in more detail on the most common interventions identi®ed in study 1.

STUDY 1 Method Participants Surveys were sent to all families (Nˆ290) receiving workshop consultations from the Multisite Young Autism Project (MYAP) in the United States on how to implement the UCLA Treatment Model (Smithet al., in press) for their 84 T. Smith and M. Antolovich Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) children with autism. All families were self-referred. Children in these families were under ®ve years old at treatment onset and had received a diagnosis of autism from a licensed psychologist or physician unaliated with MYAP. At the time of this mailing (December, 1995), MYAP sites were as follows: Northwest Young Autism Project (NYAP; Director: Tristram Smith, Ph.D.), University of California Ð Los Angeles (Director: O. Ivar Lovaas, Ph.D.), Central Valley Autism Project (Modesto, CA; Directors: Mila Amerine-Dickens, M.S., and Howard Cohen, Ph.D.), and Bancroft School (Haddon®eld, NJ; Directors: Scott Wright, B.S., and Kathy Dyer, Ph.D.). Consultations (described by Smithet al., in press) consisted of 1±3 day workshops every 2±3 months for children, their families, and paraprofessional therapists who had been hired by the families to implement the treatment.

Procedure Based on a review by Smith (1996) of interventions commonly o€ered to children with autism besides ABA treatment, a 15 minute, written survey (avail- able from the ®rst author) was developed for parents to check o€ which interventions their children had received at any point in their lives and to rate the impact of these interventions as positive, negative, or neither. The survey con- tained space for parents to write comments, if they wished, about particular interventions and to list interventions that were omitted from the survey but implemented with their child. The investigators sent survey forms and stamped, self-addressed envelopes to the directors of each site in MYAP. The directors then distributed the forms and envelopes to all families receiving consultations from their site. All forms included a cover sheet stating that the information was anonymous and that participation was voluntary. Classroom interventions were excluded from the present report because the UCLA treatment includes procedures for helping children adjust to such settings. The survey included the following interventions, which, with the exception of some speech therapies and the ®rst four biomedical interventions listed, are viewed with skepticism by most researchers (Smith, 1996).

.Non-behavior-analytic speech therapies (Prizantet al., 1997).

.Sensorimotor therapies: (a) Sensory Integration Therapy (SIT; stimulating children's skin and vestibular system through activities such as brushing parts of children's bodies, squeezing parts of their bodies, swinging on hammocks, and spinning in circles on specially constructed chairs; Ayres, 1972); (b) Auditory Integration Therapy (determining sound frequencies to which children may be sensitive and then using headphones to listen to music from Supplemental interventions 85 Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) which these frequencies have been ®ltered out; Rimland & Edelson, 1994); (c) Facilitated Communication (using an adult facilitator to guide children's hands as they type on a keyboard; Biklen, 1993).

.Individual therapies: (a) psychoanalysis ( providing as much freedom as possible in an atmosphere of warmth and love, so that children can overcome disrupted relationships with parents; Bettelheim, 1967); (b) the Greenspan (1992) model (helping children `form a sense of their own personhood' ( p. 5) through playful, emotion-laden interactions with therapists and parents); (c) humanistic play therapy ( playing with toys in a setting where children receive unconditional positive regard; DesLauriers & Carlson, 1969); Gentle Teaching (exhibiting unconditional and authentic valuing of children so that bonding occurs between children and parents; McGee & Gonzales, 1990); Options ( providing individualized, loving attention to a child in a residential setting for most of the child's waking hours; Kaufman, 1976); Holding Therapy ( forcibly holding the child so as to cause `the autistic defense . . . to crumble' (Welch, 1987, p. 48).

.Biomedical treatments: (a) antidepressants, (b) neuroleptics, (c) amphet- amines, (d) anticonvulsants, (e) anxiolytics ((a)±(e) reviewed by Campbell & Hueva, 1995), ( f ) megadoses of B6 with magnesium (Rimland, 1987), and (g) elimination diets (diets in which children are forbidden to consume additives such as preservatives or certain foods such as dairy products, wheat, or yeast; Crook, 1987). Results and Discussion Of the 290 surveys sent out, 121 (42%) were returned. Children reportedly had received an average of seven supplemental interventions (range 0±15). As shown in Table 1, the most frequently used interventions were (a) non-behavior-analytic speech and language therapy (85%), (b) megadoses of B6 with magnesium (61%), (c) Sensory Integration Therapy (56%), and (d) elimination diets (50%).

Apart from these interventions, parents reported 20 diverse, additional inter- ventions, as listed in Table 1. Individual therapies such as Gentle Teaching and standard psychotropic medications such as Selective Serotonin Reuptake Inhi- bitors (SSRIs) were rarer than sensorimotor therapies and alternative medicine.

Parents tended to rate supplemental interventions, regardless of type, as bene- ®cial (50%) or neutral (38%). They seldom deemed interventions harmful (9%).

Overall, then, children in this study were participating in numerous supple- mental interventions, and parents often rated these interventions highly. Because of the low return rate (42%), these results may not have been representative of all 86 T. Smith and M. Antolovich Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) families in MYAP. For example, parents who endorsed many supplemental treatments may have been more inclined than other parents to respond to the survey. Even so, it is evident that many families' investment in supplemental interventions was very high.

STUDY 2 The present study was designed to examine when and why parents decided to obtain supplemental interventions for their children and what speci®c bene®ts they attributed to these interventions. Of particular interest were (a) whether the onset of ABA treatment was associated with changes in the use of supple- mental interventions (e.g., whether children's participation in such interventions increased or decreased), (b) how parents learned about the interventions, (c) what Table 1. Supplemental treatments used by workshop and clinic clients at UCLA and replication sites (Nˆ121) Therapy No. of respondents Helpful Harmful Neither/unsure Speech therapy a 103 61 3 39 Sensorimotor Tx Sensory Integration 68 38 1 29 Auditory Integration 36 18 0 15 Facilitated Commun. 7 5 0 2 Psychotherapy Psychoanalysis 7 0 2 5 Greenspan Model 12 6 3 3 Humanistic Play Tx 4 1 0 3 Gentle Teaching 1 0 1 0 Options 2 0 0 2 Biomedical Tx Antidepressants b 24 11 7 6 Antipsychotics c 511 3 Amphetamines d 11 7 2 2 Elimination diets e 60 39 0 20 B6‡Mg 74 28 1 41 aSpeech therapy based on a theoretical orientation other than behavior analysis. bFluoxetine (13), clomipramine (3), imipramine (2), paroxitine (1), ¯uvoxamine (1), unspeci®ed (4).cRisperidone (1), thioridizine (1), clonidine (1), unspeci®ed (2).dMethylphenidate (7), pomadine (3), d-Amphetamine (1).e(Includes reports of multiple diets): Feingold diet (6), elimination of other foods Ð dairy (33), gluten/wheat (24), casein (8), sucrose (6), yeast (5), chocolate (4), other/unspeci®ed (11).

Note: No reports of the use of Holding Therapy or anxiolytic and anti-convulsant medications. Additional behavior analytic treatment (32); other interventions: occupational therapy (5), music therapy (4), physical therapy (2), Kaplan Visual Therapy (special eyeglasses, 2), gammaglobulin infusions (1), Cranial Vascular Therapy (1), therapeutic horseback riding (1), chiropractics/herbal therapy (1), residential treatment (1).

Supplemental interventions 87 Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) factors were most in¯uential in the parents' decision to utilize the interventions, and (d) what behavioral changes parents noted. The study focused on the four interventions that 50% or more of the respondents in study 1 reportedly used (speech therapy, megavitamins, Sensory Integration Therapy, and elimination diets).

Method Participants Participants were 24 of the 25 primary caregivers for children receiving workshop consultations from the authors' treatment site (NYAP). The criteria for enrolling children and the format of the workshop consultations were the same as in study 1. Twenty participants were the biological mothers of the children receiving workshops; 19 of these mothers were married to the children's biological fathers at the time of the study, while the other was a single parent.

The remaining four respondents were children's biological fathers (all married to the children's biological mothers at the time of the study). Seventeen mothers were homemakers; ®ve were professionals such as teachers or nurses; and two were business executives. Fathers' occupations included one homemaker, four teachers, three business executives, six professionals in engineering or comput- ing, four administrators or small business managers, two unskilled manual laborers, one skilled manual laborer, and one clerical worker. The mean number of children per household was 2.36 (SDˆ1.08, rangeˆ0±5). Thus, families were above average in the proportion of two-parent households, mothers who were homemakers, and fathers who were employed in professional occupations; they also had an above-average number of children (Berk, 2000). The mean chronological age of their children with autism was 54.09 months (SDˆ14.27).

Children had received a mean of 13.24 months (SDˆ9.56) of ABA treatment with workshop consultations from NYAP. Procedure Telephone interviews were conducted with each participant by an NYAP workshop consultant who was naõ Ève as to children's histories with supplemental interventions. The interview (available from the ®rst author) was 30±60 minutes in length. The interviewer asked a series of standardized questions about each of the four interventions (speech therapy, megavitamins, Sensory Integration Therapy, and elimination diets) on the following topics: (a) starting and stopping 88 T. Smith and M. Antolovich Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) date(s) for the intervention, (b) speci®c intervention techniques implemented, (c) the source from which parents ®rst learned about the intervention (other parents, service providers, or publications), as well as the source that was most in¯uential in their decision to obtain the intervention, (d) features of the inter- vention itself that contributed to their decisions (e.g., research support, theory or philosophy underlying the intervention, or recommendations by parents or professionals), and (e) particular problems that spurred parents to start the intervention at the particular time they did. After receiving a response to each standardized question, the interviewer had the discretion to seek clari®cation by asking individual follow-up questions. The interviewer also asked parents to rate the e€ects of the intervention for their children on a scale from 1 (much worse) to 5 (much improved), with 3 denoting no change. Parents made one rating for each of six domains: language and communication, aggression and tantrums, social skills, self-stimulatory and ritualistic behaviors, play skills, and self-help.

The interviewer wrote down responses verbatim and, after the completion of the interview, coded the responses. To check reliability, an independent rater, who was an NYAP workshop consultant, independently coded 25% of the interviews. The interviewer and rater received no training except for one hour of instruction on coding speech therapy techniques into the following categories:

.model based (e.g., Greenspan (1992) approach); .structured (speci®c activities and target behaviors for sessions selected by therapist, but no particular instructional format (e.g., discrete trials) used); .naturalistic (Wilcox & Shannon, 1998): (a) responsive adult interactions (expanding on the child's utterances and commenting on his/her activities) or (b) global interactive activities (e.g., turn taking or other activities).

Results and Discussion Interrater reliability was acceptable (96% agreement across responses).

Table 2 summarizes the types of intervention children received and parent ratings of the e€ects of these interventions. As in study 1, parents were more likely to report positive than negative e€ects. However, they almost always reported that positive e€ects were small. Indeed, there were only two reports of major gains: one parent described large improvements across behavioral domains with speech therapy, and another parent described similar improve- Supplemental interventions 89 Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) Table 2. Types of supplemental interventions and parent ratings of ecacy Speech therapy (nˆ19)Diets (nˆ7)SIT (nˆ10)Vitamins (nˆ11) Type Structurednˆ8 Global interactionsˆ3 Structured‡globalˆ3 Unknownˆ3Dairynˆ1 Wheatˆ1 Dairy‡wheatˆ2 Wheat‡yeastˆ2 Dairy‡wheat‡yeastˆ1Vestibular stimulationnˆ2 Tactile stimulationˆ2 Pressure‡vest.ˆ2 Pressure‡Tact.ˆ2 Pressure‡vest.‡tact.ˆ3B6‡Mgnˆ7 B6‡Mg‡DMGˆ4 Months of Tx (M(SD)) 15.68 (9.32) 5.43 (3.91) 8.60 (6.55) 4.00 (2.75) Amount of Tx (M(SD)) 1.19 (0.69) hours/week n.a. 1.33 (1.09) hours/week Milligrams/day:

B6: 352.50 (185.81) Mg: 195.31 (94.72) Parent ratings (M(SD)) a Language 3.55 (0.76) 3.57 (0.78) 3.27 (0.47) 3.36 (0.67) Tantrums/aggression 3.21 (0.71) 3.43 (0.98) 3.27 (0.47) 3.46 (0.69) Social skills 3.37 (0.60) 3.43 (0.78) 3.46 (0.69) 3.36 (0.67) Ritualistic behavior 3.21 (0.71) 3.14 (0.78) 3.27 (0.47) 3.55 (0.82) Play 3.21 (0.71) 3.43 (0.78) 3.18 (0.41) 3.46 (0.69) Self-help 3.21 (0.63) 3.50 (0.84) 3.18 (0.41) 3.27 (0.67) aRatings of behavioral change from 1 (much worse) to 5 (much better), with 3 denoting no e€ect. No statistically signi®cant di€erences between groups. 90 T. Smith and M. Antolovich Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) ments with megavitamins and diets. 1Seven parents who reported continuing with speech therapy, despite low ratings of ecacy, indicated that they did so because they received funding for this intervention. Two parents also cited funding as a reason for continuing SIT.

Except for megavitamins, the particular techniques used in each intervention reportedly varied across children (Table 2). Such variation could perhaps have been expected for speech therapy because service providers hold diverse theor- etical orientations (Berkell Zager, 1999) and because children with autism exhibit a wide range of communication de®cits. It may also have been predictable for diets because parents selected particular regimens based on their obser- vations of foods to which their children might be sensitive. However, the range of SIT techniques was more dicult to account for, given that service providers shared a speci®c theoretical orientation and repertoire of therapeutic techniques.

Parent interviews and reports on ®le did not indicate a clear assessment procedure for selecting particular techniques. Thus, although previous reviews of SIT have centered on the extent to which its theory and therapy are supported by research (e.g., Arendt, MacLean, & Baumeister, 1988), its assessment procedures may also warrant scrutiny. 1The report of major gains with speech therapy was made by the parent of one child who received intervention based on the Greenspan (1992) model prior to the onset of ABA treatment. This report received some corroboration from an increase in verbal IQ from 61 to 80 during this period, as measured by the Wechsler Preschool and Primary Scales of Intelligence Ð Revised (Wechsler, 1989), administered by an examiner independent of the authors's clinic. Greenspan's model emphasizes having a parent or therapist interrupt children's activities (e.g., stopping them from turning the wheels on a car) so that children must communicate with the adult in order to resume the activity. This approach could be viewed as a form of incidental teaching (Hart & Risley, 1968) and hence may be bene®cial for some children. However, the child who reportedly made large gains with this approach in the present study was concurrently receiving other services ( participating in a regular education class with a full-time aide). Hence, the e€ects of speech therapy for this child were unclear.

The one report of major gains with diets and megavitamins, which the parent introduced simultaneously to the child 5 months after the onset of ABA treatment, could not be corroborated. For example, according to records from the child's logbook for ABA treatment, the following were the number of mastered responses in all instructional programs in which multiple responses were being trained, as recorded during each workshop consultation before and after the onset of diets and megavitamins. Nonverbal Imitation Ð (B)efore: 10, 15, 37, (A)fter: 44, 44‡; Receptive Commands Ð B: 5, 8, 34, A: 38, 50; Receptive Action Labels Ð B: 1, 23, 32, A: 41, 60; Expressive Action Labels Ð B: 0, 30, 36, A: 40, 60; Receptive Opposite Pairs Ð 0, 1, 3, A: 6, 11, 28; Answering Conversational Questions Ð B: 0, 2, 14, A: 16, 16‡; Counting (one-to-one correspondence) Ð B:

10, A: 10, 15, 20. All workshop consultations occurred 10±12 weeks apart.

To determine whether an intervention might have been ecacious even though the parent reported no change, all interventions that were introduced after the onset of ABA treatment were examined. No signi®cant bene®ts were detected, when `signi®cant bene®ts' were de®ned as a 50% increase in the rate of acquisition in any ABA instructional program in which multiple responses were being trained. (An example of a signi®cant bene®t would be a 50% increase in the number of mastered responses from one workshop consultation to the next before the onset of the supplemental intervention (e.g., from 4 to 6 to 9), and a 75% increase after (e.g., from 9 to 16 to 28).) Because the analyses in this note were unplanned and uncontrolled, they must be viewed with caution. Supplemental interventions 91 Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) Table 3 summarizes parents' responses about when and why they obtained supplemental interventions. As shown, 19 of 24 reported that their children had received non-ABA speech therapy. (Four of the other ®ve, not included in the table, reported that their children had received speech therapy that was closely coordinated with their ABA treatment, though most noted that they switched therapists two or three times before ®nding one willing to work in this manner.) According to parent report, 18 of 19 began speech therapy prior to ABA treatment, with 15 of these 18 remaining in the intervention after starting ABA treatment (13 continuing to the time of the present study). By contrast, mega- vitamins, SIT, and elimination diets tended to start after ABA treatment and, with the possible exception of diets, tended to stop before the present study. Table 3. When and why parents sought supplemental interventions Speech (nˆ19)Diets (nˆ7)SIT (nˆ11)Vitamins (nˆ11) When a Before ABA onlynˆ3nˆ0nˆ2nˆ2 During ABA only 1 6 5 9 Before‡after ABA b 15140 Source of information a Medical personnel 12 0 2 0 Educators 3 0 2 0 Other professionals 3 0 6 0 Parents 1 5 0 2 Autism publications 0 2 1 9 Reason for obtaining tx a Recommendation Service provider 9 0 4 0 Parent 1 4 0 2 Author on autism 0 1 0 3 Research 0 2 0 5 Theory 0 0 4 0 Intuitive appeal 9 0 1 1 Enjoyment for child 0 0 2 0 Child problem spurring tx a Language 19 0 1 0 Disruptive behavior 0 4 7 4 Ritualistic behavior 0 1 2 0 Slow progress in ABA 0 1 0 2 None 0 1 1 5 ap50.05 for 2(speech therapy versus all other interventions combined) 2(most common response for speech therapy versus all other responses combined) chi-square. bNumber of children currently in intervention: speech (15), diets (4), SIT (3), vitamins (4).

92 T. Smith and M. Antolovich Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) Most parents indicated that they learned about speech therapy from the medical personnel who diagnosed their children with autism. However, they said they typically learned about SIT from service providers for their child (speech thera- pists, educators, and, in one case, a behavioral consultant), while they received information about megavitamins and diets primarily from other parents or from nontechnical publications on autism ( parents most often citedAutism Research Review International).

Additional di€erences emerged regarding which aspects of interventions were attractive to parents (Table 3). Professional recommendations and intuitive appeal were reportedly the main in¯uences on the decision to obtain speech therapy. Professional recommendations were also commonly cited for SIT, but never for megavitamins or diets, which were usually said to have been selected based on parent recommendations, research evidence, or advice given by writers on autism. Unsurprisingly, parents almost always indicated that delayed lang- uage was the problem that spurred them to initiate speech therapy. However, they tended to identify disruptive behaviors as the impetus for megavitamins, SIT, and diets. Across interventions, parents reported that the problems they sought to alleviate were longstanding concerns rather than acute crises.

GENERAL DISCUSSION Most children with autism in the present investigation received many supple- mental interventions in addition to ABA treatment. Parents were more likely to rate interventions as helpful than harmful, but almost always indicated that any improvements were small and isolated to a few behaviors. Factors that in¯uenced parents' decisions to obtain supplemental interventions for their children di€ered substantially, depending on the individual intervention. For example, most parents stated that they learned about speech therapy at the time their children received a diagnosis of autism, and that the source tended to be professionals involved in this diagnosis. They indicated that they did not become aware of other interventions until later, typically ®nding out about SIT from their children's therapists, about diets from other parents, and about megavitamins from nontechnical autism publications, especially newsletters.

Most children reportedly started speech therapy before ABA treatment and continued it to the time of the present investigation, whereas they tended to start the other treatments after ABA treatment and then stop them. Parents kept their children in some interventions that they considered ine€ective, reportedly because they received funding for doing so. Supplemental interventions 93 Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) The present investigation had several limitations. First, almost all families in the present study had two parents, with at least one working in a profes- sional occupation. Moreover, families were self-referred and were obtaining consultations on how to set up and run their own treatment programs in their homes, as opposed to utilizing services already available in their communities. Thus, they appeared to have above-average socioeconomic status and high motivation to secure assistance for their children. These circum- stances may have led them to initiate more interventions of all types than is usual for families in other ABA treatment programs. Also, as previously noted, the use of supplemental interventions may have appeared especially high because of the return rate in study 1 (42%). In addition to these possible biases, the reliance on parental reports in studies 1 and 2, as well as the small sample for some interventions in study 2, may have yielded information that was sometimes unreliable. Finally, because of the limitations of such reports, it was not possible to assess the quali®cations of those who implemented supplemental interven- tions or the ®delity with which these interventions were applied.

Despite the foregoing shortcomings, however, the present investigation revealed extensive use of supplemental interventions by one group of children with autism in ABA treatment. Moreover, this use appeared to increase rather than decrease after the onset of ABA treatment. Such increases have also been reported in other clinical populations receiving a variety of interventions (Lam- bert & Bergin, 1994, pp. 175±176). Investigators have interpreted this phenom- enon as evidence that individuals were reinforced for seeking help and hence were predisposed to obtain additional help. Such an interpretation may apply to parents in the present investigation. However, the parents themselves took a di€erent view, reporting that they started supplemental interventions after ABA treatment mainly because they did not ®nd out about the interventions until then.

As with many other interventions for behavior problems (Seligman, 1995), parents more often rated supplemental interventions favorably than unfavor- ably, though they generally described bene®ts as modest. Controversy exists over the extent to which such favorable ratings re¯ect genuine improvement (Seligman, 1995), as opposed to response bias (e.g., motivation to describe an intervention as worthwhile after having devoted time, e€ort, and resources to it; Jacobson & Christensen, 1996). In the present study, parent reports appeared grossly consistent with data from behavior observations, except for one uncorroborated report of large improvements (Footnote 1). To obtain more detailed information on the correlation between parent report and other evalu- ations, future research should focus on interventions that, in contrast to those investigated in this study, have clearly de®ned, speci®c, objectively measured behavioral goals. 94 T. Smith and M. Antolovich Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) The frequent use of supplemental interventions and the tendency to rate these interventions favorably indicate that, despite their often questionable research basis, supplemental interventions have high social validity (Schwartz & Baer, 1978; Wolf, 1978). Moreover, this social validity appears to be a function of multiple antecedents and consequences. Antecedents cited by parents included professional referrals, parent recommendations, and information suggesting that interventions had scienti®c support. Consequences included not only perceived bene®ts from the interventions but also other favorable outcomes such as increase funding for their children.

Given the prevalence of supplemental interventions, treatment outcome studies on children with autism may need to include measures of such inter- ventions. Studies to data have not done so (cf. Smith, 1999). Also, because most of these interventions were unsupported by methodologically rigorous research (New York State Department of Health, 1999) and were described by parents as providing little or no bene®t, it seems important to reduce their use. In view of the multiple antecedents and consequences associated with such interventions, this may be a complex undertaking. For example, professionals have often recommended that investigators combat questionable interventions by reporting their ®ndings in the popular media and rebutting credulous stories that appear in such outlets. This strategy might be e€ective with interventions such as mega- vitamins, for which nontechnical publications on autism were reportedly the primary in¯uence. However, it might not be e€ective with other interventions such as diets or SIT, for which person-to-person recommendations rather than the popular media were most frequently cited. Educating parents and pro- fessionals on how to evaluate whether treatment recommendations are scienti®c- ally sound may help with these interventions. However, education by itself may be insucient. Parents in the present study often reported that they kept their children in interventions, despite having concluded that the interventions were unhelpful, because they received funding to do so. Besides education, therefore, it may be necessary to in¯uence the policies of funding agencies. In general, then, professionals are likely to need a multifaceted approach to reduce the use of supplemental interventions. ACKNOWLEDGEMENTS This research was supported by NIMH grant MH 48663 (Multisite Young Autism Project). Portions of the research were presented as a poster at the annual meeting of the Association for Behavior Analysis, San Francisco, CA, 1996. Supplemental interventions 95 Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) REFERENCES Aman MG, Van Bourgondien ME, Wolford PL, Sarphare G. 1995. Psychotropic and anticonvulsant drugs in subjects with autism: prevalence and patterns of use.Journal of the American Academy of Child and Adolescent Psychiatry 34: 1672±1681.

Arendt RE, MacLean WE, Baumeister AA. 1988. Critique of sensory integration therapy and its application to mental retardation.American Journal on Mental Retardation 92: 401±411.

Ayres AJ. 1972.Sensory Integration and Learning Disorders. Western Psychological Association:

Los Angeles.

Berk LE. 2000.Child Development, 5th edn. Allyn and Bacon: Boston.

Berkell Zager DE (ed.) 1999.Autism: Identi®cation, Education, and Treatment, 2nd edn. Erlbaum:

Mahwah, NJ.

Bettelheim B. 1967.The Empty Fortress. Free Press: New York.

Biklen D. 1993.Communication Unbound: How Facilitated Communication is Challenging Traditional Views of Ability/Disability. Teachers College: New York.

Campbell M, Hueva JE. 1995. Psychopharmacology in child and adolescent psychiatry: a review of the past seven years. Part 1.Journal of the American Academy of Child and Adolescent Psychiatry 34: 1124±1132.

Charlop-Christy MH, Kelso SE. 1997.How to Treat the Child with Autism. Claremont, CA.

Crook WG. 1987. Nutrition, food allergies, and environmental toxins.Journal of Learning Disabilities20: 260±261.

DeMyer MR, Hingtgen JN, Jackson RK. 1981. Infantile autism reviewed: a decade of research.

Schizophrenia Bulletin 7: 388±451.

DesLauriers AM, Carlson CF. 1969.Your Child is Asleep: Early Infantile Autism. Dorsey:

Homewood, IL.

Green G. 1996. Early behavioral intervention for autism: what does the research tell us? In Behavioral Intervention for Young Children with Autism, Maurice C (ed.). Pro-Ed: Austin, TX; 29±44.

Greenspan SI. 1992. Reconsidering the diagnosis and treatment of very young children with autistic spectrum or pervasive developmental disorder.Zero to Three 13:1±9.

Hart B, Risley TR. 1968. Establishing the use of descriptive adjectives in the spontaneous speech of disadvantaged preschool children.Journal of Applied Behavior Analysis 1: 109±120.

Jacobson JW, Mulick JA, Schwartz AA. 1995. A history of Facilitated Communication: science, pseudoscience, and nonscience.American Psychologist 50: 750±765.

Jacobson NS, Christensen A. 1996. Studying the e€ectiveness of psychotherapy: how well can clinical trials do the job?American Psychologist 51: 1031±1039.

Kaufman BN. 1976.Son-Rise. Harper and Row: New York.

Lambert MJ, Bergin AE. 1994. The e€ectiveness of psychotherapy. InHandbok of Psychotherapy and Behavior Change, 4th edn, Bergin AE, Bergin SL (eds). Wiley: New York; 143±189.

Matson JL, Benavidez DA, Compton LS, Paclawskyj T, Baglio C. 1996. Behavioral treatment of autistic persons: a review of research from 1980 to the present.Research in Developmental Disabilities 17: 433±465.

McGee JJ, Gonzales L. 1990. Gentle Teaching and the practice of human interdependence: a preliminary group study of 15 persons with severe behavior disorders and their caregivers. In Perspectives on the use of Non-Aversive and Aversive Interventions for People with Developmental 96 T. Smith and M. Antolovich Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) Disabilities, Repp AC, Singh NN (eds). Sycamore: Sycamore, IL; 215±230.

New York State Department of Health 1999.Clinical Practice Guideline: the Guideline Technical Report. Autism/Pervasive Developmental Disorders, Assessment and Intervention for Young Children. Publication 4217. Author: Albany, NY.

Prizant BM, Schuler AL, Wetherby A, Rydell P. 1997. Enhancing language and communication development: language approaches. InHandbook of Autism and Development Disorders, 2nd edn, Cohen DJ, Volkmar FR (eds). Wiley: New York; 572±605.

Rimland B. 1987. Megavitamin B6 and magnesium in the treatment of autistic children and adults. InNeurobiological Issues in Autism, Schopler E, Mesibor GB (eds). Plenum: New York; 390±405.

Rimland B. 1998. UCLA ®ndings: intensive intervention helps even low-functioning children.

Autism Research Review International 12(1): 1±2.

Rimland B, Baker SM. 1996. Alternative approaches to the development of e€ective treatments for autism.Journal of Autism and Developmental Disorders 26: 237±241.

Rimland B, Edelson SM. 1994. The e€ects of Auditory Integration Training on autism.American Journal of Speech±Language Pathology 3: 16±24.

Schwartz IS, Baer DM. 1991. Social validity assessments: is current practice state of the art?

Journal of Applied Behavior Analysis 24: 189±204.

Seligman MEP. 1995. The e€ectiveness of psychotherapy: TheConsumer Reportsstudy.American Psychologist 50: 965±974.

Smith T. 1996. Are other treatments e€ective? InBehavioral Treatment for Young Children with Autism, Maurice C (ed.). Pro-Ed: Austin, TX; 45±59.

Smith T. 1999. Early intervention for children with autism.Clinical Psychology: Science and Practice 6: 33±49.

Smith T, Donahoe P, Davis BJ (in press). The UCLA Young Autism Project. InPreschool Education Programs for Children with Autism, 2nd edn, Handleman JS, Harris SL (eds). Pro- Ed: Austin, TX.

Wechsler D. 1989.Wechsler Preschool and Primary Scales of Intelligence. (rev.). Psychological Corporation: San Antonio, TX.

Welch MG. 1987. Toward prevention of developmental disorder.Pennsylvania Medicine90:

47±52.

Wilcox MJ, Shannon MS. 1998. Facilitating the transition from prelinguistic to linguistic communication. InTransitions in Prelinguistic Communication, Wetherby AM, Warren SF, Riechle J (eds). Brookes: Baltimore, MD; 385±416.

Wolf MM. 1978. Social validity: the case for subjective measurement, or how behavior analysis is ®nding its heart.Journal of Applied Behavior Analysis 11: 203±214. Supplemental interventions 97 Copyright# 2000 John Wiley & Sons, Ltd.Behav. Intervent.15: 83±97 (2000) Copyright of Behavioral Interventions is the property of John Wiley & Sons, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.