Naturalistic Developmental Behavioral Interventions as Value-Based and Culturally Adapted EBPs for Autistic Individuals

Abstract

Naturalistic Developmental Behavioral Interventions (NDBIs) are child-centered and motivation-based interventions for autistic children. Though they hold great promise due to their naturalistic implementation and focus on children’s strengths, there have been recent calls to improve NDBIs such that they are more aligned with the neurodiversity approach. Central to this argument is the notion that autistic clients and their families should find the intervention acceptable. This chapter describes how NDBIs differ from other behavioral interventions, what their strengths are, and how they can continue to be improved. More specifically, we focus on expanding target NDBI outcomes to include more distal, family centered variables (such as family quality of life); improving the social validity of NDBIs; and how NDBIs can be implemented using a tiered support system.

1. Introduction

A steady increase in the prevalence of autism across the globe and growing demands for service and support have drawn increasing attention from policy makers, researchers, and practitioners in the field of health care, mental health care, and disability services. A recent systematic review, synthesizing 99 estimates from 71 studies of the prevalence of autism spectrum disorder (ASD) worldwide, suggests that approximately 1% of children are diagnosed with ASD around the world and that the prevalence data in 2012 and 2021 are not much different from a global point of view [1]. However, there are a number of countries (e.g., Australia, France, Korea, and the USA) reporting a noticeable ascending trend of autism prevalence. For instance, in the USA, the prevalence of autism has been rising, and it is rising very rapidly according to the CDC. When the CDC started systemic screening in 2000, the ratio of 8-year-old children with ASD was about 1:166, and it has increased fourfold to 1:44 as of 2020 [2]. In response to the increasing autistic population and huge demands for high-quality service and support, there have been intensifying efforts in researching and developing effective intervention approaches for supporting autistic individuals to improve specific outcomes such as social communication and interaction skills and enhance holistic outcomes like quality of life. Over the last several decades, a wide array of psycho-educational interventions has been developed and researched, leading to the need for a systematic way of determining which of these interventions are most effective for autistic individuals. As such, the field has identified its own methods for identifying what we know as evidence-based practices (EBPs), which are seen as the “gold standard” in the field of autism intervention.

In this chapter, we will discuss current EBPs in the autism intervention literature. We begin with an introduction of the most commonly identified EBPs and a brief summary of some major methodological and ethical concerns of these psych-educational approaches. In the pursuit of potential remedies for addressing these issues and concerns, we introduce Naturalistic Developmental Behavioral Interventions (NDBIs; [3]), as a highly regarded intervention model with a strong evidence base, and potential value-based and culturally adapted evidence-based practices that can help address some of the underlying methodological and ethical concerns in autism interventions and support. We will then delineate both promises and limitations of NDBIs. Lastly, we evaluate the areas in which NDBIs need to be further improved, as well as implications for clinical practice and future research.

2. Evidence-based practices in autism interventions and support

There is a plethora of research on autism interventions that have been conducted in the last five decades. This research has sought to answer a critical question: how do we effectively support autistic individuals and their families? As a result of this research, the field has been able to identify a set of interventions that have proven to be effective and demonstrated positive outcomes for autistic individuals. The field knows these as evidence-based practices (EBPs). Currently, there are two main groups that have evaluated the extensive literature to determine which interventions and intervention packages qualify as evidence-based practice—the National Standards Project [4] and National Professional Development Center (NPDC) on autism spectrum disorders. Through a thorough literature review, NSP has identified 14 established evidence-based practices and 18 with emerging evidence [4]. The NPDC identified 27 evidence-based practices [5]. Most of these identified EBPs are based on behavioral principles.

Based on the NPDC standards, in order to meet the criteria as an evidence-based practice, there must be positive evidence to support the intervention strategy with at least two high-quality experimental or quasi-experimental group design studies that have been conducted by two different researchers or research groups, or at least five high-quality single-subject design studies conducted by at least three different researchers or research groups, totaling at least 20 participants. A practice could also meet this standard if there is one high-quality randomized or quasi-experimental group design study and a minimum of three high-quality single-subject design studies conducted by at least three different researchers or research groups across all studies [5]. The NSP standards for established evidence-based practices are similar and require either two group designs or four single-subject design studies that have a minimum of 12 participants with no conflicting results or a minimum of three group designs of six single-subject designs with a minimum of 18 participants, with no fewer than 10% of studies having conflicting results. Additionally, the NSP has a scoring system for ensuring these studies are of high quality [4].

Even with these standards, there have been criticisms of evidence-based practices used with autistic children from both methodological and ethical perspectives (e.g. [6]). These issues identified are not only with the literature base, but also with the practical implications of the use of EBPs in clinical settings. In terms of methodological issues, there have been concerns with both the quality of the review methods for determining the evidence-based interventions for autistic children, as well as issues with the quality of individual study designs and protocols. Specifically, Sandbank et al. [7] identified three major concerns with the reviews of the National Standards Project (NSP) and National Professional Development Center (NPDC) on Autism Spectrum Disorders: the attempt to blend evidence from different research design methodologies (single-subject and group designs) though there is no agreed upon way of doing so, limited information about the extent of intervention effectiveness and for which populations, and the lack of summarizing the effects of any given intervention. In addition to these clear issues with synthesizing the literature on autism interventions, there has been a multitude of common methodological issues that are seen across individual studies. Specifically, critics have expressed concerns over the strong focus on single-subject design versus group design and randomized controlled trials (RCTs) as well as the focus on short-term outcomes, with little attention paid to long-term effects [8]. On a similar note, an additional concern is that there is a lack of focus on adverse outcomes in the literature on autism interventions [9, 10]. Finally, there are concerns about who is conducting the research. This comes with the lack of autistic involvement in autism research, the lack of research on autistic input on evidence-based practices, and the limited social validation measures used in the autism intervention literature [11, 12, 13]. Moreover, researchers and research groups who are conducting this intervention literature often fail to identify any potential conflicts of interest (COI), though they are often relevant, calling into question the potential for bias within these studies [14].

Even more concerning than the methodological shortcomings are the ethical concerns that have been voiced by autistic self-advocates in recent years. We have identified a number of ethical concerns that have been brought up regarding the implementation of commonly used evidence-based practices for autistic children. Chief among these concerns are: the unethical history of behavioral intervention, the focus on normalization of autistic recipients, and the emphasis on compliance (a more detailed discussion is provided in a later section regarding limitations of Naturalistic Developmental Behavioral Interventions).

2.1 Naturalistic developmental behavioral interventions

One example of a manualized EBP approach for autistic children is Naturalistic Developmental Behavioral Interventions (NDBIs; [3, 15]). NDBIs are a set of intervention models that combine behavioral principles with insights from developmental psychology. Intervention models considered as NDBIs include pivotal response treatment (PRT; [16]), Early Start Denver Model (ESDM; [17, 18, 19]), Project ImPACT [20], incidental teaching (IT; [21]), enhanced milieu teaching (EMT; [22]), and Joint Attention Symbolic Play Engagement and Regulation (JASPER; [23]). All NDBIs emphasize child-centered learning focused on meeting the child where they are and using their interests to improve their motivation to learn [3]. NDBIs have built a large evidence base of both single-subject studies and randomized controlled trials, though methodological issues such as publication and detection bias and reliance on parent report measures [7, 24] weaken the evidence base. Nonetheless, a recent meta-analysis still found that NDBIs had the most promising evidence base compared to other interventions for autistic children [7].

Implementation of NDBIs should occur in natural environments, throughout the child’s daily routines, at home, or at school. They were originally in part created to combat some of the issues clinicians saw with more typical, structured ABA, such as lack of motivation and generalization [25]. As such, all NDBIs share common components such as following the child’s lead, shared control, natural reinforcement, and reinforcing non-perfect attempts. For example, if a child loves cars, playing with toy cars might be used to teach words such as “car,” “go,” or “fast.” The child’s interest in cars would thus motivate them to practice such words in a fun, enjoyable environment. It is also important in NDBI implementation to have shared control between children and adults, as opposed to more structured ABA, where the clinician usually directs interactions and has control over the materials [26]. This shared control gives the child agency but also allows the adult the chance to create learning opportunities (for example by holding out the car and asking the child, “what color is this?”). Natural reinforcement is also a hallmark of NDBIs. Natural reinforcement differs from external reinforcement in that the reinforcement is the natural consequence of whatever the child just did. For example, if the child wants to play with a car and says, “car,” or answers the question about the color of the car, the adult contingently reinforces them with the car (not with candy, a sticker, or a token). Lastly, NDBIs encourage clinicians to reinforce attempts, as opposed to pushing for fully correct responses. For instance, a child might approximate the word car by saying “cah” or they might say a car is “pink” when it is actually red. By reinforcing these non-perfect attempts, children learn that it is the effort that matters to the adults in their lives.

3. Maximizing family intervention fit: cultural adaptations, neurodiversity, and strengths-based approaches

Family involvement is always a core aspect of NDBIs. Parents should be involved in goal setting as well as the intervention implementation; in fact, to maximize the use of the natural environment, most NDBIs specifically incorporate parent training such that parents can act as intervention providers [27]. However, simply involving parents in discussions about goals and teaching them to implement NDBIs does not guarantee that the interventions are meeting the families’ needs. Therefore, researchers and clinicians must ensure that intervention recipients are involved in ways that are meaningful to them, whether that is more or less involved or involved in different ways. We present three areas that could lead to improvements in family intervention fit: adapting interventions to fit different cultures, incorporating a neurodiversity approach, and emphasizing strengths-based approaches.

3.1 Culturally adapted EBPs

Cultural adaptations of EBPs, which have emerged as an area of study in other fields (e.g., clinical psychology, see Lau [28]), seem to be absent as a focus of EBP discourse in the context of delivering quality service and support to autistic students for better life outcomes. In particular, a fundamental assumption about the universal applicability and effectiveness of EBP to all populations in the field of mental health care and service appears to be a flawed one when evidence of EBP effectiveness is solely based on the results of certain populations (e.g., White students of suburban middle-class families) [29]. Likewise, there seem to be similar issues about the legitimacy of implementation and dissemination of EBPs to populations of color in special education considering that most EBPs have been developed without even taking account of the cultural context of ethnic minority communities and cultural identity of those presumably being served (e.g., autistic individuals).

Castro and colleagues define the concept of cultural adaptation as: “a planned, organized, iterative, and collaborative process that often includes the participation of persons from the targeted population for whom the adaptation is being developed” ([30], p. 215). Cultural adaptation also concerns the process of adjusting an evidence-based intervention protocol by taking language, culture, and context into account to make it compatible with the cultural patterns, meanings, and values of those being served [31]. Culturally adapted EBPs for autistic individuals can help not only clarify and specify what ought to take place in adaptation for obtaining an optimal balance between adaptation and clinical implementation fidelity that leads to important implementation outcomes such as acceptability and appropriateness, but also define and clarify the necessary knowledge, skills, and roles of those facilitating the implementation process and the timeline and sequence of adaptations in the implementation process [32].

3.2 What is neurodiversity?

Many critiques of behavioral intervention can be tied to the fact that applied behavioral analysis (ABA) usually uses the medical model of disability to “treat” autism. Under this view, autism is seen primarily as a medical issue, with “reduction of autism symptoms” and appearing more “normal” as intervention targets. This is in direct contrast to the model of neurodiversity, which sees autism as a different—but not inherently bad—type of brain wiring. In her writing, Judy Singer [33], the sociologist who coined the term “neurodiversity,” described it as similar to biodiversity, in that different types of brains contribute to the ecological fitness of a society. Harvey Blume’s essay in The Atlantic, which introduced the term to the public in 1998, also used the word in this way: “Neurodiversity may be every bit as crucial for the human race as biodiversity is for life in general. Who can say what form of wiring will prove best at any given moment?”

Today, the term “neurodiversity” is used in multiple different ways [34]. The first is similar to the way Singer [33] and Blume [35] described neurodiversity: as a biological fact. Everybody’s brain is different, which by default implies that there is diversity in human neurology (hence, neuro-diversity). The second way “neurodiversity” is often used refers to the ways this biological fact is utilized in everyday life (known as the neurodiversity paradigm (Walker [34] or neurodiversity approach [36]). According to Walker [34], there are three fundamental aspects of the neurodiversity paradigm/approach: neurodiversity is a beneficial form of diversity; there is no “average” or “normal” brain; and social dynamics affecting other forms of diversity (such as gender and race) are also at play when it comes to neurodiversity. Dwyer’s [37] breakdown of the neurodiversity paradigm/approach is similar: there are no ideal minds; neurodivergent people should be accepted, not normalized; and society plays a large role in disabling neurodivergent individuals. Dwyer [36] also highlights that many advocates have adopted their own personal definitions of neurodiversity. For example, Bailin’s [38] definition of the neurodiversity approach is inherently political: “primarily a call to include and respect people whose brains work in atypical ways, regardless of their level of disability.” All of these definitions of neurodiversity relate back to autistic activist Jim Sinclair’s [39] Do not Mourn For Us speech (see [40] for a historicization of this speech) where Sinclair implored parents to accept their autistic children and to stop seeing autism as a tragedy, explaining that autism is not something that can (or should) be removed from an individual. They highlighted that autism is a way of being (not an appendage or a “shell”) and that autistic individuals are not stuck behind an impenetrable wall, incapable of building relationships with others. Sinclair acknowledges that parents may need to grieve the loss of their idea of an “ideal” child, but having an autistic child in and of itself is not a reason to grieve. This emphasis on acceptance of difference is the backbone of neurodiversity.

3.3 Common misconceptions about neurodiversity

Though neurodiversity hinges upon the biological fact of differences in brains, many misconceptions arise when the neurodiversity approach is applied to individuals with disabilities such as autism. Den Houting [41] brings up three common misconceptions. First, some critics feel that the neurodiversity approach views autism and other neurological disabilities as simply differences and not disabilities, implying that this approach effectively erases disability (e.g., [42]). Advocates highlight that this is just simply untrue—while the neurodiversity approach does see neurological variation as differences that should be respected, they still acknowledge the potentially disabling properties of such differences [38, 41, 43, 44, 45] (den Houting, 2018). While some contend that the disabling aspects are solely due to the social environment (e.g., [41]), others reject a strict social model and recognize that impairments localized in the individual can themselves be disabling (e.g., [37, 43]).

Another common critique is that the neurodiversity approach really only applies to those who have higher IQ and who can verbally self-advocate (e.g., [42]; numerous posts on the National Council for Severe Autism’s blog, e.g., [46]). Den Houting [41] argues this ignores the fact that ability profiles of autistic individuals are not always consistent, or they might be great at something 1 day but struggle with it the next. Therefore, to assume that someone who can speak verbally is all-around “high-functioning” is incorrect. Ballou [43] and den Houting [41] also highlight that some neurodiversity self-advocates are in fact minimally verbal autistics (though den Houting acknowledges that there is a need for greater representation of non- or minimally speaking individuals within the movement). Ballou also makes the point that even if some individuals cannot traditionally self-advocate verbally, their other communicative acts should still be listened to and respected.

The last critique discussed by den Houting [41] is that, because neurodiversity advocates call for acceptance, they think neurodivergent people do not need any support. While this is not espoused by nearly any proponent of the neurodiversity paradigm, den Houting [41] sees this critique as stemming from a disagreement regarding the outcomes of support services. Those who support the neurodiversity approach call for supports that emphasize quality of life (e.g., [47]) and accommodations (e.g., [48]), whereas its critics tend to utilize the medical model wherein autistic individuals should be remediated and normalized. It is therefore a mischaracterization to say that neurodiversity proponents do not want disabled individuals to receive supports; they just want supports in place that are accepting and validating of one’s neurotype.

3.4 Strength-based approaches to interventions/supports for autistic individuals

Though many early interventions for autistic individuals use the medical model to understand and “intervene upon” autism, there are researchers who feel that the neurodiversity paradigm and early intervention can coexist, particularly if clinicians listen to autistic perspectives regarding intervention goals, procedures, and outcomes [49, 50, 51, 52]. Therefore, understanding neurodiversity is crucial for professionals who work with autistic people, as it can help the field move away from a focus on deficits and normalization and toward figuring out what autistic people actually want and need. One way for interventionists and support professionals to embrace neurodiversity in their practice is to use strengths-based approaches. Such approaches recognize that everyone, no matter what challenges they might face, has strengths and that such strengths must be acknowledged and leveraged in education, support services, and employment to ensure individuals enjoy a high quality of life.

In typical intervention/education for autistic children, deficits are identified through standardized tests and questionnaires, and plans such as individualized education programs (IEPs) are created in order to remediate these problem areas. These plans typically leave little room to document strengths [53], implying either that disabled children do not have strengths or that their strengths do not matter in comparison to their deficits. For example, a child’s joint attention abilities may be overshadowed or even ignored if the child often needs to run and jump around the room while listening, as their IEP might call for them to sit still. Furthermore, some challenges faced by autistic individuals can be viewed as strengths depending on the context [54]. For example, perseverance on specific topics is often perceived as a problem, though it could be reconceptualized as excellent focus and expert knowledge. Autistic individuals have been shown to have a multitude of different strengths [55, 56], and capitalizing these interests and passions can be a way to motivate individuals to learn things they may have not been interested in doing previously [57, 58].

4. Promises and limitations of NDBIs

Despite cumulative evidence of the effectiveness of EBPs for autistic children suggested by several systematic literature reviews [4, 5], there are a great number of methodological and ethical concerns raised in the more recent meta-analytic review by adopting more sophisticated research methods with respect to the evidence base of EBPs in general (e.g., behavioral interventions) and NDBIs specifically [7, 14, 24]. NDBIs, emerging as valued-based approaches, have begun to address many of these methodological and ethical concerns in the implementation process of interventions and support for autistic children. NDBIs exhibit numerous promises in promoting strengths-based approaches, centering autistic voices, and eventually improving efficacy and effectiveness of interventions and support for autistic individuals in naturalistic settings, and also shed light on the future landscape of culturally adapted and value-based EBPs for helping and supporting autistic individuals to achieve better life outcomes.

4.1 Promises of NDBIs

NDBIs hold many promises, not just in their encouraging efficacy [7], but also in their philosophical underpinnings. As discussed above, NDBIs excel in their strengths-based approach to teaching children, focusing on what they can do and not what they cannot. Emphasizing child choice over clinician-directed tasks helps boost child motivation, which allows for a more positive, fun intervention environment. By demonstrating to children that effort is more important than perfection, children build confidence to try new things.

The use of naturalistic intervention settings and natural reinforcement also addresses some of the shortcomings of more traditional EBPs. By learning that good things happen during natural interactions, children are able to more easily generalize these skills across environments. For example, a child who learns during an NDBI intervention session that requesting a favorite toy result in getting that toy will find that that same behavior results in the same natural reinforcement at school. Furthermore, there is no need to constantly ensure that providers are all using the same external reinforcers across settings (e.g., making sure that teachers are using the same crackers at school as the interventionist uses at home), since the (natural) reinforcement is built into all learning opportunities. All these above-mentioned attributes of NDBIs attest to its strong embracement of the ecological perspective of child development and its applications in establishing child-centered and home-based interventions since NDBIs are typically implemented in autistic children’s natural contexts (e.g., home and community settings) through everyday routine interactions with caregivers. This will likely improve the social and ecological validity of interventions and support for autistic children. In addition, NDBIs hold promises as developmentally appropriate practices which are broadly adopted in early childhood education and service for all children with exceptional learning needs given that the learning targets and objectives of NDBIs are guided by early developmental sequences. NDBIs also have the capability of being highly culturally responsive, as they emphasize setting goals that are important to the family [59, 60] and family professional partnership [61]. This emphasis on family thus should include intervention considerations related to culture. Several studies of culturally adapted NDBIs have shown promise. For example, researchers in China adapted the ESDM curriculum to emphasize vocal tones and include examples of eating relevant to Chinese society (e.g., pictures of chopsticks instead of forks) [62]. In India, Project ImPACT was adapted based on initial parent feedback. Changes included adding more psychoeducation about autism, teaching parents about play, and including extended family members as intervention providers [63]. Additionally, several NDBI manuals have been translated into languages other than English, such as Italian [64] and Chinese [65] versions of the ESDM manual.

Cultural context is also important when considering NDBIs’ potential alignment with the neurodiversity approach. Some in the autistic community see themselves as part of an autistic cultural community that is distinct from other groups [66]. Just as NDBIs have the capability to be culturally adapted to fit other ethnic or racial groups, NDBIs have the potential to be culturally adapted to be aligned with autistic culture [52].

In contrast to most of the evidence-based behavioral interventions which predominantly focus on proximal outcomes of behavioral changes and skill acquisition, NDBIs cast emphasis on broader goals and long-term outcomes (e.g., social inclusion, rights, emotional, physical, and material well-being, self-determination, and quality of life). For example, PRT aims at training autistic children in their pivotal areas of motivation, self-initiation, and self-management that can lead to effective learning of both proximal outcomes such as cognitive, social, and other functional and life skills and distal outcomes like self-determination and quality of life.

4.2 Limitations of NDBIs

Apart from numerous promises that NDBIs can bring to improve outcomes of interventions and support for autistic children, we also acknowledge some limitations of NDBIs. Similar to those generic methodological issues identified with evidence-based practices in autism interventions above, NDBIs exhibit methodological limitations on the issues such as the quality of the review methods for determining the intervention effectiveness (e.g., primarily synthesizing the effects of the different interventions using a narrative approach and lack of using meta-analytic methods) and the quality of study designs and protocols (e.g., evidence base of NDBIs is primarily backed up by a research literature of plenty of single subject design studies and limited number of RCTs and rigorous group experimental design studies). NDBI RCTs are also hindered by the same methodological issues that autism intervention studies face more generally [24].

Another shortcoming in autism intervention studies is missing conflict of interest (COI) statements [9]. Chief among the potential COI issues are: (1) the author was the intervention model developer; (2) the author is employed to provide the intervention or is affiliated with an institution that provides said intervention; (3) the author receives direct payments for services or items related to the intervention (e.g., training materials, books, providing training); and (4) the study uses a commercially available measure that was developed by the author [9]. Lack of transparency in reporting these conflicts can cast doubt on the validity of the evidence base of the NDBIs interventions. It is important that researchers should adequately report any potential conflicts of interest in the NDBIs intervention studies.

Even more concerning than the methodological shortcomings are the ethical concerns that have been voiced by autistic self-advocates in recent years regarding the implementation of commonly used interventions for autistic children. Three major complaints stand out: the unethical history of behavioral intervention, the focus on normalization of autistic recipients, and the emphasis on compliance. While NDBIs in theory have attempted to address some of these problems, some of these concerns remain relevant to these more naturalistic approaches. More specifically, NDBIs have sought to improve upon the early iterations of behavioral interventions, focusing on reinforcement instead of aversive punishments.

However, goals of intervention, regardless of if it is through the use of NDBIs or more structured, traditional behavioral approaches, are often based on neurotypical standards defined by non-autistic people. This has been linked to autistic masking or camouflaging, the phenomenon that autistic people often feel the need to mask their autism traits in an effort to fit in (see [67, 68]). What is most concerning is that masking has been linked to poor mental health outcomes [69]. Though the implementation of NDBIs is not directly associated with the act of suppressing autistic traits, the common goals of these practices are. For instance, teaching vocal language or social conversational skills based on neurotypical norms are common goals of these interventions. In fact, most NDBI research focuses on improving “social communication” and “language” skills [24], with very few NDBI studies focusing on using alternative communication modalities [70].

The other major ethical concern with NDBIs is the issue of compliance [71]. This goes hand in hand with the focus on normalization, as autistic children are taught compliance around neurotypical behaviors, such as engaging in specific conversational behaviors, or forced eye contact, both of which have been known to be targets of NDBIs. Though there is no doubt that NDBIs seek to provide an increase in autonomy for children participating in the intervention (through the use of child choice/following a child’s lead), the use of reinforcement to get children to do certain things or engage in certain behaviors can inadvertently place too much emphasis on compliance, even in these more gentle, naturalistic approaches. As such, NDBI researchers and clinicians need to pay attention to the practical ways in which they are using NDBIs to ensure that the client always maintains their autonomy and that skills being taught are meaningful to the client, so that quality of life and autonomy are prioritized over compliance.

Yet another concern lies with the positionality of the researchers. Historically, it has been neurotypical persons conducting research on interventions and delivering interventions for autistic people, with little involvement from the autistic community. This issue is relevant across all intervention research, including that of NDBIs. The evidence base for all interventions would be significantly strengthened with the increase of autistic researchers involved in these studies, as well as an increase in research and clinical practices that seek autistic feedback on intervention methods and practices to determine which practices are even found to be appropriate and effective by autistic people themselves.

The last concern on NDBIs intervention studies has to do with its historical focus on short-term outcomes, and lack of attention to long-term effects [7]. Though immediate effects of an intervention are of obvious importance, it is concerning that there have not been more longitudinal studies. This concern is of particular importance in light of the voiced concerns from autistic self-advocates that interventions for autistic children have resulted in trauma [72]. Moreover, given that autistic individuals are at a greater risk for mental health issues such as anxiety and depression in adulthood [73], there is a great need to evaluate the long-term effects of these interventions being deemed as “effective” in the short-term to ensure that long-term outcomes are just as positive. On a similar note, an additional concern is that there is a lack of focus on adverse outcomes in the literature on autism interventions [10, 14]. Just because something is found to be effective in terms of identified outcome measures, it does not mean that it is without adverse effects on its participants, and as such it is equally important to specifically evaluate these in all intervention studies.

In addition, it is noted that in systematic reviews and quantitative syntheses, positive outcomes proximal to intervention targets are often reported as evidence of intervention effectiveness for autistic children [24, 74]. Few studies conduct follow-up measures of child and family outcomes beyond 3–6 months after interventions and support are delivered to autistic children and their families [8]. Long-term outcomes of interventions for autistic adults are even more concerning. Some research results seem to portray a troubling picture on the long-term outcomes of adults with autism suggesting that adults with ASD have rather limited social integration, poor job prospects, and high rates of mental health problems and that overall outcomes for autistic adults in the areas of jobs, relationships, independent living, and mental health are considerably poorer than their same age neuro-typical peers [75]. A recent meta-analysis of studies of quality of life (QOL) for individuals with ASD across the lifespan suggests that autistic adults including those with higher intellectual and verbal functioning have poorer QOL than their peers without ASD [76].

5. Implications for future research and practice

Sandbank et al. [7] and Crank [24] both point out that, while NDBIs are a promising intervention type, evidence from randomized controlled trials indicates that the evidence base relies too heavily on parent report, which could inflate effect sizes. This must be addressed in intervention research. However, two important notes must be made: (1) parents may actually be able to detect subtle clinically significant changes in their children that are not picked up on by more standardized measures; and (2) even a “placebo-by-proxy” effect could lead to positive outcomes [77]. For example, parents might be highly encouraged after an NDBI trial because they saw improvements in their child’s socio-communicative abilities that they had not witnessed before. Perhaps they then slightly overestimate their child’s language improvements on a post-study questionnaire. While it is true that the effect size may be inflated by the parent’s self-reported data, it is also possible that this family will continue to implement the intervention because they are so happy with it, which could lead to further language improvements. That said, future research can still do a better job of accurately capturing parent-reported outcomes. For example, qualitative methods such as interviews or free-response questions could be used to corroborate quantitative questionnaire data and explore why parents might be overestimating certain things. Furthermore, new questionnaires should be specifically developed to measure parent-reported changes in child behavior, as it is generally not considered psychometrically valid to use instruments in ways in which they were not originally intended and designed [78, 79]. Instead of using instruments designed as clinical diagnostic tools (e.g., the Vineland Adaptive Behavior Scales [80] or the Social Responsiveness Scale [81]), instruments must be specifically designed and validated to measure within-person, longitudinal change in parents’ perceptions of their child’s abilities.

Though context-bound outcomes can be useful [7, 82], researchers must ensure that they are also including outcome measures that evaluate generalization. For example, many PRT studies rely on parent–child interaction videos to evaluate changes in communication (e.g. [59, 60, 83, 84]). Though parents are not necessarily always instructed to implement PRT (e.g., Hardan et al. say that “parents were instructed to try getting the child to communicate as much as possible” (p. 886)), it is likely that parents knew they were instructed to use PRT, especially since these same videos were used to assess PRT fidelity, and the majority of parents did indeed meet fidelity. While these video clips are important in that they demonstrate that children are responsive to PRT while it is being implemented, it is not clear that improvements in socio-communicative skills would necessarily generalize to other environments where PRT was not being directly implemented, such as at home with their family or at school. Therefore, we suggest that other observational measures be included, such as interactions with adults not involved in or trained in the intervention, interactions with other children such as siblings or classmates, and/or natural interactions at home (e.g., see [85] for a discussion of using 16-h at-home recordings to analyze vocal reciprocity).

In addition to the importance of improving outcome measures of autistic children’s generalized gains from interventions and support, there seem pressing needs to incorporate more distal outcomes of autistic individuals such as: changes of autism core characteristics, changes of the implementation process (e.g., person-centered and family centered planning that emphasizes preferences of autistic individuals and their families), and environment (e.g., sustainable family routine), and holistic measures of improved relationship and satisfaction with enhanced well-being (e.g., family and professional partnership and individual and family quality of life) in intervention effectiveness studies. More research needs to be conducted by including longer follow-up assessments of all above mentioned distal outcomes of autistic individuals and their families. Furthermore, future research ought to examine how different aspects of multicomponent intervention programs like NDBIs affect different types of outcomes. Additionally, adopting recent intervention research models like the Sequential Multiple Assignment Randomized Trial (SMART; [86]) in the future research of NDBIs can help researchers and interventionists not only systematize the application of a personalized approach to NDBIs [87] but also investigate the course of change for adapting the intervention processes to autistic children who are viewed either as a treatment “non-responder” or as “hidden victim” of adverse effects of the intervention [52]. By taking an additional measure to ensure the validity of NDBIs, we can benefit from employing participatory action research in the future research of NDBIs in which autistic individuals and their families are included in the process at all stages of the clinical trial or clinical program development.

5.1 Implications for practice

First, increased uptake of NDBIs is needed throughout the clinical community. Regrettably, many behavior interventionists have a limited understanding of NDBIs [88], and community implementation is lacking [89]. Further training is thus needed in how to implement NDBIs effectively. Furthermore, it is important for clinicians to understand the need to actively consider cultural adaptations, neurodiversity, and emphasizing strengths when working with autistic children and their families. Crucially, these should not be after-thoughts that are only addressed when NDBIs are implemented outside the United States or when a family points out that something is not in line with their values. Clinicians should bring these topics up from the beginning, regardless of the client’s cultural background, in order to establish that these are meaningful aspects of the intervention that are taken seriously by the provider.

Clinicians must also always take autistic perspectives into account. As suggested by Schuck et al. [52], this may mean hiring autistic consultants or behavior analysts, or if that is not possible, staying up to date on what the autistic community is saying via academic literature and social media. This also applies to the perspective of the autistic intervention recipients themselves. If recipients are old enough and can communicate via speaking or writing, interviews or questionnaires can be administered to assess how clients are feeling about the intervention. For younger clients and/or those who use other means of communication, other types of treatment acceptability measures should be used (e.g., assessing child affect [90]).

Furthermore, intervention goals need to be neurodiversity affirming. That is, goals need to be based on autism acceptance and should not encourage children to go against an autistic way of being [52, 91]. Feedback from autistic adults can help clinicians and researchers develop more neurodiversity-friendly goals. For example, pushing children to exclusively use spoken language over other communication alternatives is seen by many to be pushing neurotypical standards on autistic children who have different communication needs (Schuck et al., under review). Additionally, autistic adults view many social intervention goals (e.g., improving communication skills; learning skills of conversation) skeptically (Baiden et al., under review). Relatedly, clinicians must always center autistic clients’ strengths. Emphasizing strengths can improve children’s self-esteem and confidence, lead to skill generalization, and ultimately lead to improvements in quality of life.

Apart from ensuring the appropriate and respectful goals of the interventions, it is crucial for both researchers and clinicians to examine if the implementation of NDBIs is appropriate and respectful and the intensity of NDBIs is appropriate as well. Autistic individuals and their families as well as professionals alike often expect to receive some common intervention recommendations (e.g., intervention variety and intensity). For example, it is noted that intensive behavioral interventions, delivered at 25–40 h per week, are the most frequently recommended intervention for young children with autism. However, Sandbank et al. [9, 14] contend the notion that “greater intervention intensities were associated with greater intervention gains” (p. 341). Research evidence shows that NDBIs, often provided at an intensity ranging typically from 10 to 20 h per week, can help autistic children achieve significant gains in multiple targeted and untargeted areas of interventions.