روندها و مباحث در مداخلات رفتاری ویژه اولیه برای کودکان نوپای مبتلا به اوتیسم

The use of applied behavior analysis (ABA) to treat persons with autism goes back several decades. Many specific target behaviors and intervention strategies have been developed. In the last two decades the most heavily studied of these methods has been Early Intensive Behavioral Interventions (EIBI). This package of ABA methods is unique in two ways. First, a broad range of target behaviors are trained for 20–40 h per week. This training is much more treatment per week than what is described in most ABA studies. Second, the children treated are typically 2–3 years of age, which is younger than for most ABA research. Reviews of EIBI have typically focused on the efficacy of the methods. These are important, but at present we argue that these methods are effective. This paper is different in that it looks at current trends such as generalization, parent training, factors that mitigate against effective treatment and the need for follow-up and booster treatment.

Autism is an area receiving intense attention from researchers worldwide. Genetics are the most frequently researched of all topics. However, other topics are also studied often (Matson and LoVullo, 2009a and Matson and LoVullo, 2009b). For example, applied behavior analysis (ABA) has become a well established treatment approach (Kodak et al., 2009, Strachan et al., 2009, Sturmey et al., 2005 and Williams, 2010). This set of assessment and treatment techniques is based on operant conditioning, and has been especially popular for persons with developmental disabilities (Matson et al., 1999b, Matson et al., 1999c and Matson et al., 1998). Perhaps the most visible and frequently studied of these disorders are autism, PDD-NOS, and Asperger's Syndrome (Brim et al., 2009, Matson et al., 2008a, Mayes and Calhoun, 2009, Reilly, 2009, Young et al., 2009 and Zalla et al., 2009). These three conditions are often referred to as autism spectrum disorder and in DSM-V, will all be under the diagnosis of autism. These disorders are neurodevelopmental in origin and symptoms are evident early in life (Gillberg, 2010, Leung et al., 2010, Nyden et al., 2010 and Peters-Scheffer et al., 2010). The triad of communication, social skills, and rituals and stereotypies are part and parcel of this disorder (Matson et al., 2009b and Njardvik et al., 1999). These problem areas have been traditional topics of study in the field of applied behavior analysis (Adcock and Cuvo, 2009, Dixon et al., 2010 and Matson et al., 2009e). Also common in the autism spectrum are challenging behaviors (CB) (Coe et al., 1999 and Duncan et al., 1999; Kuhn & Matson, 2001; Matson et al., 1997, Matson and Rivet, 2008, Paclawskyj et al., 1997, Paclawskyj et al., 1997, Paclawskyj et al., 2001, Rojahn et al., 2003, Smith and Matson, 2010a and Smith and Matson, 2010b). As with the core features of autism, CB are frequently treated with ABA (Applegate et al., 1999, Matson et al., 1999a and Matson et al., 2008b). Psychopathology is also a comorbid disorder that occurs frequently with autism and often is confused with CB. This problem had led to over medication, and the application of medications for problems where they will not prove to be effective, and can compound existing problems, specifically with respect to serious side effects (Fodstad et al., 2010 and Matson et al., 2010). This problem may largely be alleviated by the development of more effective ABA methods within both assessment and treatment. A review of these and related issues follows, with particular emphasis on current trends and future developments. 1.1. Early Intensive Behavioral Interventions (EIBI) 1.1.1. Why empirical methods are important Arguably the most effective treatment for autism is applied behavior analysis. Because of this, intervention at very young ages, generally by 2 or 3 years, is recommended. Some researchers have even gone on to conclude that ABA is the only treatment that produces comprehensive, lasting results for autism (Foxx, 2008). Developing evidence based methods is essential, and it is incumbent on professionals to educate the general public on these methods. Tremendous amounts of misinformation are available, and are exacerbated by ill informed and/or unscrupulous professionals who advocate for all manner of bogus treatments (Shute, 2010). And, the internet, while a wonderful development in many respects, makes it very easy to disseminate untested, nutty treatments. For example, Green et al. (2006) identified 111 different treatments for autism. They surveyed 552 parents of children with autism and found that on average seven different treatments had been tried for every child. Many of these treatments have limited research to support their use such as special diets (27%), vitamin supplements (43%) and sensory integration. Unfortunately, this is not all about selecting treatments based on good data. In a particularly telling study, Schreck and Mazur (2008) surveyed 469 Board Certified Behavior Analysts (BCBA). Not all BCBA supported the use of ABA methods to treat autism, although most were so inclined. Even more amazing, and at the same time disappointing, BCBAs endorsed and used all types of untested treatments despite their belief that the treatments were difficult to implement, not cost effective and not supported by research. This study is important since it suggests that market forces that go above and beyond demonstrated treatment effectiveness are driving treatment decisions. BCBAs would be among the groups with the greatest knowledge of EIBI. The fact that this group is not uniformly on board with EIBI as a first treatment option does not speak well for wide spread dissemination. Having said that, treatment effectiveness should be the primary rationale for choosing an intervention. 1.1.2. Empirical support Peters-Scheffer, Didden, Korzilius, and Sturmey (2011) conducted a meta-analysis of 11 studies with 344 children who had autism. They conclude, as have many others, that the results strongly support the effectiveness of EIBI (Hayward, Eikeseth, Gale, & Morgan, 2009). Also, Goin-Kochel, Myers, Hendricks, Carr, and Wiley (2007) found high rates of impairment but none of the children were able to enter school without additional staff support. Similarly, Shi, Yu, Guo, and Li (2007) did a follow-up for 48 children from an initial group of 85 who at ages 2–6 years received 30–40 h of EIBI for 3–12 months. They found that 43 of 48 children continued to improve, with 29 entering normal kindergarten. Granpeesheh, Tarbox and Dixon (2009), based on studies such as this, concluded that there is a significant amount of research evidence supporting EIBI. Remarkably, despite thousands of ABA-EIBI studies on specific core deficits, and related challenging behaviors and skills, and EIBI studies as well, some researchers still question the efficacy of these methods. Morris (2009) argues that this occurs because some researchers misunderstood and/or misrepresent ABA. One of the most recent of these papers is by Spreckley and Boyd (2009) who conclude that ABA did not produce better outcomes than standard care for children with autism. This conclusion and the errors in the paper are breathtaking. Equating one package method, EIBI, to the vast number of ABA methods that have been developed is one stark example. Smith, Eikeseth, Sallows, and Graupner (2010) provide an excellent rebuttal of their study. Smith et al. (2010) however stress that they agree with Spreckley and Boyd (2009) that large multi-element randomized clinical trials are needed to narrow the confidence interval for effect sizes. We do not support this approach. Studies on ABA and EIBI, whatever the level of methodological sophistication, routinely demonstrate good or superior effectiveness. The effects of all studies are in the direction of positive effects. From our perspective this issue is no longer debatable, and has not been for some time. 1.1.3. Cure or not Lovaas (1987) went so far as to suggest that these interventions cure autism, at least for some children. Given that these markedly improved children have not been followed into adulthood, such a claim cannot be substantiated. Granpeesheh, Tarbox, Dixon, Carr, and Herbert provide a more measured assessment of EIBI effectiveness. They conclude that a subset of children achieve a level of functioning that is indistinguishable from typically developing peers. Cure implies that regression in these skills will not occur even without additional treatment. The more likely scenario is typical functioning can be achieved in some autistic children, but is likely to regress without additional, periodic intervention. Houlin (2010) also makes some excellent points. She states that EIBI have the best empirical support to date. Houlin (2010) also underscores the point that large individual differences occur across children. She stresses the need to determine which components work best for specific individuals and under which conditions other researchers have also made this point (Alessandri, Thorp, Mandy, & Tuchman, 2005). Granpeesheh et al. (2009) underscore the conceptual issue that EIBI should not be decouple when it is discussed in the professional literature from the vastly more extensive ABA on which EIBI is based. As these authors have noted, hundreds of research papers have demonstrated the efficacy of ABA. Positive reinforcement, stimulus control, shaping, fading, chaining, functional assessment and generalization are some examples. These techniques have proven to be effective across a host of age groups, disorders and problems. Thus, discrete trial training, when described as the major/primary component of EIBI is a distraction. In fact, this is just one of many ways to deliver ABA. It is not a principle law of learning. Rather, it is a strategy, one of many, to deliver reinforcement and behavior decelerators. Thus, while Discrete Trial Training (DTT) is useful, no research has been conducted to demonstrate that this system for ABA is superior to other delivery methods.

The area of EIBI has become perhaps the most promising research area in the field of autism. The most unique aspects of the interventions are the large number of hours of treatment weekly (20–40), and the very young age of the children treated (2–4 years). While a small minority of professionals still denies the efficacy of these methods, the vast majority of researchers who have examined the evidence are of the opinion that improvements occur in varying degrees for all children who participate. Lovaas and associates have contributed to this literature. However, to refer to such methods as Lovaas therapy is unwarranted. Azrin, for example, demonstrated the effectiveness of massed practice for children and the developmentally disabled with toileting problems a decade before EIBI emerged. Baer, Risley, Iuata and many others have contributed substantially to the development of core ABA methods and principles used in EIBI such as shaping, chaining, generalization, prompting and functional assessment. This point is made not to detract from any given researcher's contribution, but to point out that EIBI draws from an entire, large and robust field of research (ABA). Thus, thousands of studies are embodied in the outcomes of EIBI versus the contributions from a few studies. Assuming the point that EIBI works, it is now time to move on to related points. Many researchers apparently agree since efforts to determine methods of treatment integrity, generalization and maintenance, and how to tailor intervention to children who display this heterogeneous condition. These are very important developments. Professionals and parents are likely to see a good deal more movement of this sort in the near future. Given the general acknowledgement that starting EIBI as early as possible couples these interventions to early diagnosis. Major advances have been made in this area in recent years as well. Another issue related to this one is the evaluation of treatment outcomes. We have consistently been skeptical of claims that IQ could be dramatically increased. First, at young ages, until 6 or 7 years of age, standard IQ tests are notoriously unreliable. Second, marked improvements in IQ scores are likely due to increased compliance and attention. Additionally, tests designed to evaluate core symptoms of autism, frequently occurring challenging behaviors and comorbid psychopathology have infrequently been assessed at post test (Matson, 2007a and Matson, 2007b). Not only is this critical, but the test used should be normed on the age group treated. More refinements of procedures and even more effective treatments are likely.