A growing number of alternative therapies claim positive outcomes for children with ASD (Schreibman 2005; Schreck and Miller 2010; Smith and Antolovich 2000), but applied behavior analytic interventions have the most empirical evidence of effectiveness for treating ASD and at this time are the only evidence-based interventions (National Research Council 2001; Schreibman 2000; Sherer and Schreibman 2005). Therefore, best practices for treatment following diagnosis include EIBI, or comprehensive applied behavior analytic intervention delivered to children 5 years old and younger (see Peters-Scheffer et al. 2011 for a review of comprehensive EIBI programs). Unlike focused interventions, which are implemented for a limited period of time to improve specific targeted behaviors (Hall 2009; Machalicek et al. 2007, 2008), comprehensive applied behavior analysis (ABA) models of intervention center around intensive intervention (typically 20–40 h each week for 2 or more years). Intervention consists of carefully structured, massed teaching trials or discrete trial training (DTT; Leaf and McEachin 1999; Lovaas 1981) and/or naturalistic ABA intervention approaches including incidental teaching (e.g., Hart and Risley 1975), pivotal response training (PRT; e.g., Koegel et al. 1987, 1989) ; and enhanced milieu teaching (e.g., Kaiser et al. 2000; Kaiser and Hester 1994). Communication and social skills, play, self-help and independent living skills, challenging behavior, cognition and preacademic skills are often targeted with consideration of typical child development (Leaf and McEachin 1999; Vismara and Rogers 2010). Additionally, some EIBI programs first teach foundational or readiness skills such as joint attention, compliance, attending/orienting to stimuli, choice making, and imitation (e.g., Frea and McNerney 2008). Currently, EIBI models that are procedurally well described, have been replicated, and have some evidence of efficacy include or are associated with: (a) the Denver Model, (b) Learning Experiences: An Alternative Program for Preschoolers and Parents (LEAP), (c) the Lovass Insititute, (d) May Institute, and (e) the Princeton Child Development Institute (see Odom et al. 2010 for a review of comprehensive EIBI models). Across each of these models, well-prepared clinicians supervise program development and delivery and rely on significant parent involvement. Although researchers continue to debate the number of required hours and methodology of intervention, and the targeted skills and curriculum, there is general consensus that intensity of the program contributes significantly to positive child development, (Eldevik et al. 2006) as does parent participation (Sallows and Graupner 2005). Empirical evaluations of the effectiveness of EIBI programs have included both clinic-based (Ben-Itzchak and Zachor 2007; Eikeseth et al. 2000; Fenske et al. 1985; Harris et al. 1991) and home-based programs (Anderson et al. 1987; Birnbrauer and Leach 1993; Eikeseth et al. 2002; Howard et al. 2005; Lovaas 1987; McEachin et al. 1993; Sheinkopf and Siegel 1998). Although center-based programs encourage and often require some level of parent participation, for logistical reasons, parents are generally more actively involved in home-based programs.

There are a number of benefits of involving parents in EIBI. As the primary caregivers, parents can improve the quantity and quality of intervention, their involvement can improve the generalization of positive educational outcomes to home and community settings, and their participation may decrease stress and improve coping strategies. Moreover, active collaboration with parents in assessment and the selection of goals and objectives, intervention procedures, and outcomes aligns with family-centered practices in early intervention (Dunst et al. 1994) and may improve the contextual fit of interventions and contribute to improved treatment adherence (Hieneman and Dunlap 2001). This section reviews intervention studies involving parents as part of an EIBI program.

Parent-directed EIBI programs may result in outcomes similar to clinic-based EIBI programs (Sallows and Graupner 2005). Sallows and Graupner’s experimental group design study compared the outcomes of a clinic-directed program and a parent-directed program for 24 children with autism, who were assigned to the two groups through a matched random process. Children in the clinic-based group received an average of 38 h of intervention a week and the parent-directed group received 31.5 h of intervention a week (one child received 14 h a week) with less frequent professional supervision. Somewhat surprisingly, children in both groups showed similar, positive improvements in adaptive behaviors, language, social skills, academics, and IQ. Sallows and Graupner suggested that the positive results might be explained by parents’ motivation and willingness to ensure high quality intervention (e.g., filling cancelled therapist shifts themselves, arranging play dates, taking on the senior therapist role).

However, a larger number of researchers have reported less than optimal progress when parents, rather than experienced behavior analysts manage the child’s EIBI program (Bibby et al. 2002; Smith et al. 2000b; Smith et al. 2000c). Bibby et al. (2002) evaluated the effects of parent-initiated approximations to the UCLA workshop model (Lovaas 1996) that included significant parent participation on the developmental outcomes of 66 children with ASD using an accelerated multicohort longitudinal design (Kazdin 1998). Participating children began EIBI at a mean age of 45 months, and at the time of the first assessments in this study, the majority of the children were in their second year of intervention. Children received a mean of 30.3 h of therapy each week with parents managing aspects of their child’s therapy, but the extent to which paid therapists and parents directed and/or implemented intervention was unknown. Standardized assessments, parent interview, and direct observations of the child in their family home were conducted twice, approximately 12 months apart. Adaptive behavior gains were identified in 33 % of children and 27 % experienced significant IQ gains. However, among the findings of this study was that none of the 42 children that were at least 72 months of age and had received 2 or more years of intervention had outcomes meeting Lovaas’s (1987) criteria for “best outcome.” Bibby et al. (2002) suggested that pretreatment variations in participating children, or the quantity and quality of received intervention might have contributed to these outcomes. There is growing evidence to support an explanation of differential responding to intervention due to individual child characteristics (Ben-Itzchak and Zachor 2007; Howlin et al. 2009; Sherer and Schreibman 2005). Additionally, variation in EIBI program implementation (e.g., quality and frequency of supervision) is common (Love et al. 2009).

Although parents have long been involved in or directed their child’s EIBI program, we know little about the impact of participating in EIBI programs on family functioning (Grindle et al. 2009; Howlin et al. 2009). The findings of recent studies examining the effects of mothers’ participation in EIBI suggest that many parents will desire some level of participation in EIBI, find EIBI beneficial (Boyd and Corley 2001; Dillenburger et al. 2004; Grindle et al. 2009), and are unlikely to suffer serious negative consequences from participating (Birnbrauer and Leach 1993; Hastings and Johnson 2001; Smith et al. 2000b). These findings do not discount the stress of daily experiences that families participating in EIBI encounter, such as the potential loss of privacy due to therapists regularly entering their home; and difficulties obtaining funding, managing therapists, and administering the program (e.g., Cattell-Gordon and Cattell-Gordon 1998; Grindle et al. 2009). However, for many families, the stress of participating in EIBI may not be more than the day-to-day stress they experience as a parent of a child with an ASD. For instance, Hastings and Johnson (2001) found that mothers participating in EIBI reported similar amounts of stress to mothers of children with autism who were participating in other research studies.

Although the above parent training procedures are often effective, the range of symptom presentation in ASD, level of parent education and experience, and other pragmatic issues (e.g., time parents have available to implement intervention) would seem highly likely to influence the success of parent training and parent-implemented intervention (Matson et al. 2009). Therefore, additional research focused on the factors that influence the successful training of parents and the effects of parent-implemented intervention is warranted. For example, Lang et al. (2009) reviewed research in which parents were trained to implement communication interventions to children with autism. Studies were excluded from the review if the parents’ ability to implement specific intervention components was not measured in baseline. This ensured that improvement in parent ability was measured in at least a pre/post test or AB design. The result of focusing only on studies with baselines of parent behavior was that only 11 studies involving only 60 parents qualified for inclusion in the review. Of that group, none of the studies involved a component analysis to determine the contribution of individual training procedures, and an absence of parent demographic information was noted across studies. Similar issues regarding research designs and the absence of parent participant characteristic information have been reported in other literature reviews covering parent-implemented social, communication, and challenging behavior interventions (Matson et al. 2009; McConachie and Diggle 2007; Meadon et al. 2009; Schultz et al. 2011). Therefore, definite statements regarding the most efficient and effective approaches to training parents to implement interventions to autism are not yet possible and additional research in this area remains warranted.