As a person trained in research, and a member of a neurodiverse family, I tend to place a critical eye on current ‘gold standard’ approaches to thinks like therapies. In recent years, I have seen a number of autistic adults speaking out against Applied Behavior Analysis (ABA) as a therapy. In looking critically at the research in this light, I am finding mounting evidence that Applied Behavior Analysis and related therapies such as Early Intensive Behavioral Intervention (EIBI) and Positive Behavioral Interventions and Supports (PBIS) are not the ‘evidence-based’ therapies that they claim to be. In fact, they may lead to harm. The evidence I see leads to five arguments questioning ABA:

1. The ‘evidence base’ is actually very weak. There are very few well-designed, rigorous and unbiased studies.
2. The effectiveness of the therapy has not been proven definitively. If you look at the few well-designed, rigorous and unbiased studies that do exist, they are inconclusive concerning the effects of ABA.
3. ABA uses a functional behavior assessment that is biased towards assuming that a behavior is related to a deliberate choice that a child is making and against the idea that the behavior may be a result of distress. This bias leads to the therapist operating from false assumptions.
4. There are no long-term psychological safety studies around the use of ABA, though recent studies have shown a correlation between ABA and the risks of Post-Traumatic Stress Disorder (PTSD) and suicide.
5. Ethicists are raising concerns about the ethics of ABA practices particularly in the areas of self-determination, consent, and harm.

This combined evidence brings me to the point of questioning whether it is appropriate for us to be offering and supporting ABA therapy in behaviorally-challenged children.

In the following sections, I will be covering the research in more detail. This article is somewhat academic as it is examining the research evidence. Also, within all quotes, all instances of emphasis were added by me.

Argument 1: The evidence base is weak

There is no rigorous, verifiable basis for the long-term effectiveness of ABA in experimental evidence — despite claims to the contrary. An evidence base emerges when a therapy is shown to be effective in well-designed, rigorous studies with objective metrics and a sufficiently large sample size of both autistic children and in a control group. Furthermore, the experiment needs to be repeatable — if later people try to do the same study with a different sample set the results of this new experiment should be in line with the results of the original experiment. Finally, the study should be forthright about any conflicts of interest that exist with respect to any author.

The meta-study here [1] is one of the few meta-studies that takes a close look across the broad range of autism-related research studies. They look specifically for good research practices like the use of randomized controlled trials (RCTs).

They state,

“We found that when study quality indicators were not taken into account, significant positive effects were found for behavioral, developmental, and NDBI intervention types. … [W]hen effect estimation was limited to RCT designs and to outcomes for which there was no risk of detection bias, no intervention types showed significant effects on any outcome. [1]

We note that this meta-study looked at several interventions in young children, including but not limited to behavioral interventions.

There is a second meta-study here [2], from Cochrane. This meta-study looks specifically at RCT and controlled trials related to EIBI in children between 30 and 43 months old. They found ‘no evidence at post-treatment that EIBI improves autism symptom severity’ and also ‘no evidence at post-treatment that EIBI improves problem behaviour’. They considered the evidence presented in the different ABA studies they looked at to be very low quality.

Argument 2: The best evidence brings into question the effectiveness of the therapy

In addition to the evidence mentioned above, there are others who have examined the outcomes in children who have received ABA. One of the more interesting set of reports are the analyses of the effectiveness of ABA therapy funded by TRICARE, the insurance agency associated with the US Department of Defense. They analyze annually the tens of thousands of health insurance cases related to ABA therapy for autistic individuals. In their 2020 report [3] the US Department of Defense basically stated that we have no baseline to understand how autistic children grow and develop without ABA, and thus we have no way to show that any observed improvements are actually a result of ABA. Doing nothing and waiting may be an equally effective strategy, we just cannot tell from the data.

While the change scores demonstrated small but statistically significant improvements after 12 and 18 months of rendered ABA services, and that most baseline severity scores and most ages demonstrated some percent change in scores from baseline, there was no comparison group (no treatment or another type or of treatment) to determine the attribution of these changes. It is also not clear if these changes are clinically significant. Subsequently, there is no way to know if the relatively small change observed here is the result of ABA services, other services received, or if this simply a result of maturation. However, the findings are clear that the number of hours of ABA services rendered did not improve symptoms of ASD based on the PAC scores. This finding strongly suggests that the small changes noted are not related to ABA services. [3]

The 2021 report [4] reaffirmed that there is no clarity over whether or not ABA leads to significant improvement. In fact, in an uncomfortably large percentage of the recipients of ABA, there was either no change or things got worse.

The findings from this analysis demonstrate that some beneficiaries have made some statistically significant improvements (57 percent), while other beneficiaries show no improvement or even worsening of symptoms (43 percent) over the two-year period. However, clinical significance is still unknown. [4]

Finally, this well-executed randomized controlled trial [5] looks at PBIS interventions for behaviorally-challenged adults, and comes to very disappointing results: ‘Staff training in PBS, as applied in this study, did not reduce challenging behaviour.’

Argument 3: Behavioral intervention is biased towards starting with bad assumptions

When a child starts ABA, their therapy is developed based on a Functional Behavior Assessment (FBA). The overall goal of looking for where a behavior comes from is embedded in the nature of the FBA. However, this benefit is overshadowed by the assumption that the antecedent to the behavior is rooted in the desire either to avoid something unpleasant, or to obtain something desirable (e.g., see here [6]). If this is as far as the analysis drills down to, then the resulting intervention does not attack the root of the problem. There may be a very good reason why the child is avoiding something, for instance, like the child is trying to avoid an abusive situation but is not in a strong enough psychological state to explain that. Or a child may want something because of an unmet need that the parent does not understand.

In other words, ABA operates on the assumption that the child’s behavior is a choice they are making deliberately, but downplays the possibility that the child’s behavior is emerging involuntarily from real distress.

ABA therapy becomes problematic when the therapist’s approach is based on false assumptions, and therefore does not address an autistic child’s behavioral problems at their roots. For example, this blog [7] concludes:

In many schools, PBIS is based on a flawed model that assumes challenging behavior is intentional when, often, a child is experiencing a stress response. [7]

The question then is, where is the stress response coming from? Possibilities include such things as lagging skills in emotional regulation and executive function, and sensory system related issues.

Lagging Skills

In this article [8], the authors examine the correlation between lagging skills and challenging behaviors in autistic children (youth with ASD) and no intellectual disabilities (ID). They write,

[O]ur findings suggest that children with lagging skills in emotion regulation and executive function are more likely to exhibit challenging behaviors. [8]

They also conclude:

Based on our preliminary findings, and assuming a causal relationship between the risk factors identified here and challenging behaviors, treatment for challenging behaviors in school-age youth with ASD without ID [intellectual disability] may need to specifically target the child’s emotion regulation and executive function difficulties. [8]

Targeting the development of these lagging skills looks beyond the behaviors, and thus is a very different approach than that taken by ABA. In fact, skills like executive function are better addressed by someone like an Occupational Therapist.

Sensory System

Another situation that may be underlying an autistic child’s behavior is related to their sensory system and the fact that it may be over- or under- reactive. We tend to drastically underestimate the sensory challenges of autistic children and write off their valid distress. This article [9] discusses the relationship of sensory over-responsivity (SOR) with stress and anxiety. They write:

The findings provide support for the theoretical model of this relationship in which SOR associated with autistic traits leads to stress and anxiety … Findings suggest a role for the introduction of sensory neutral environments within schools, workplaces, and other extreme sensory environments to reduce anxiety symptoms often associated with high autistic traits. … sensory problems can have serious and widespread negative effects on an individual’s life and wellbeing. [9]

We note that one of the go-to interventions for sensory over-responsivity in ABA is extinction (See, for example, here [6]), which is basically teaching the child to tolerate the overreaction. The student is to stay in the situation of sensory overwhelm and learn not to misbehave in that setting. However, this report suggests the opposite approach of modifying the environment for the comfort of the child is to be preferred.

Argument 4: Psychological studies bring into question the long-term safety of ABA

ABA has not been shown to be safe, long-term. Most studies seem to look forward at most 18 months or so. There is very little evidence from a long-term perspective. Recent studies however show a correlation between ABA and the long-term risks of risk of PTSD and suicide. In this research survey[10], the author concludes,

Nearly half (46 percent) of the ABA-exposed respondents met the diagnostic threshold for PTSD, and extreme levels of severity were recorded in 47 percent of the affected subgroup. Respondents of all ages who were exposed to ABA were 86 percent more likely to meet the PTSD criteria than respondents who were not exposed to ABA. Adults and children both had increased chances (41 and 130 percent, respectively) of meeting the PTSD criteria if they were exposed to ABA. [10]

While this published research was based on a survey approach, and was somewhat biased towards responses from females, this result is concerning.

A second safety red flag relates to the fact that one of the goals of ABA is to teach the autistic child to repress natural but ‘inappropriate’ behaviors and replace them with ones that are more ‘appropriate’. In other words, ABA effectively teaches masking/camouflaging, which is the ability to disguise or suppress specific autism traits or behaviors in social situations.

Unfortunately, there is increasing evidence that masking leads to poor mental health and suicide. For example, see this article [11].

Evidence for higher suicide rates in autistic people, and particularly among women, also is found here [12]. They conclude,

In this nationwide retrospective cohort study that included 6 559 266 persons aged 10 years or older living in Denmark during the period from 1995 to 2016, individuals with a diagnosed autism spectrum disorder had more than 3-fold higher rates of suicide attempt and suicide compared with all other persons after adjusting for sex, age, and time period. [12]

They also state,

Another important issue is the belief that autistic individuals should camouflage or mask their autistic traits to conform with societal expectations, for example by forcing themselves to make eye contact with others even when doing so is uncomfortable. … Quantitative and qualitative research studies show that such camouflaging is exhausting and is associated with poor mental health, including suicidal thoughts and behavior. This has important implications for many interventions, including social skills training and behavioral therapies that aim to normalize appearance and behavior at the risk of exacerbating a disconnect between the true self and performing self, potentially increasing anxiety and decreasing self-esteem. [12]

Argument 5: Ethicists are questioning whether ABA is ethical

The way ABA is practiced today, especially with children, is that the parent and therapist decide on what behavior goals to address, and how to address them. There is rarely any place for the child to consent to those goals, or to provide their feedback during the goal-setting process. ABA practitioners emphasize intensity, so the child is subjected to therapy without their consent for many hours a week. This removes both the child’s self-determination and their time to play and do things they enjoy.

This article [13], published in the Kennedy Institute of Ethics Journal, takes a close look at the practices and goals in Applied Behavior Analysis. The major ethical breech that they see is that ABA removes the autonomy of the autistic person being treated, from a misplaced desire to help them fit in. They state,

… we have argued that autistic children should not be treated with (the dominant species of) ABA, as such treatment essentially violates their autonomy, and at least contingently it does them direct harm and is unjust to one group of people. We have argued that the demand to help people does not make ABA obligatory, and so on balance it is to be avoided. [13]

The authors stress that ABA does not take into account the well-being of the autistic recipient of the therapy, but rather focuses on changing their behavior to conform to a more typical social structure. They also make the interesting ethical observation,

from a social perspective, we would do well to structure our society in such a way that people who see the world differently can fit in without having to either change who they are or act like someone they’re not. [13]

In other words, we should not force them to adapt to our way of thinking; rather, we should learn to understand and accept them, and take on some of that burden if adapting ourselves.

This article [14] addresses the ethical implications of intensive ABA therapy on non-speaking autistic people. The authors write,

Psychologists, like various other professionals, are charged with the responsibility to do no harm, and to “safeguard the welfare and rights of those with whom they interact professionally” (American Psychological Association, 2010, 2017). Considering their continued support, promotionand implementation of ABA treatment in the nonverbal population, how could one argue the negative effects of ABA are congruent with this oath?
In sum, … it is evident that we need to pause and look at what has resulted. … Compliance, learned helplessness, food/reward-obsessed, magnified vulnerabilities to sexual and physical abuse, low self-esteem, decreased intrinsic motivation, robbed confidence, inhibited interpersonal skills, isolation, anxiety, suppressed autonomy, prompt dependency, adult reliance, etc., continue to be created in a marginalized population who are unable to defend themselves. [14]

These papers clearly show that the ethics of ABA, and its practice in light of the oath to ‘do no harm’, are being questioned actively among ethicists.

Conclusion

In this article, I have brought together the published research that questions the underlying research, the effectiveness, the assumptions made when defining a behavior plan, the long-term psychological safety and the ethics of Applied Behavior Analysis. This underlying evidence compels me to question the use of Applied Behavior Analysis as an ‘evidence-based’ intervention, and to be concerned that the harm it may be doing exceeds any benefit the child may get from receiving it.

In part 2 of this series, I will be discussing some of the things I have been learning about the mind and the nervous system — how we think and react — and how these learnings lead me to question the effectiveness of ABA as well.

References

[1] Sandbank M, Bottema-Beutel K, Crowley S, Cassidy M, Dunham K, Feldman JI, Crank J, Albarran SA, Raj S, Mahbub P, Woynaroski TG. Project AIM: Autism intervention meta-analysis for studies of young children. Psychol Bull. 2020 Jan;146(1):1–29. doi: 10.1037/bul0000215. Epub 2019 Nov 25. PMID: 31763860; PMCID: PMC8783568.

[2] Reichow B, Hume K, Barton EE, Boyd BA. Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD). Cochrane Database of Systematic Reviews 2018, Issue 5. Art. No.: CD009260. DOI: 10.1002/14651858.CD009260.pub3

[3] The Department of Defense Comprehensive Autism Care Demonstration Annual Report 2020: Reprot on Efforts Being Conducted by the Department of Defense on Applied Behavior Analysis Services.

[4] The Department of Defense Comprehensive Autism Care Demonstration Annual Report 2021: Reprot on Efforts Being Conducted by the Department of Defense on Applied Behavior Analysis Services.

[5] Hassiotis, A., Poppe, M., Strydom, A., Vickerstaff, V., Hall, I., Crabtree, J., . . . Crawford, M. (2018). Clinical outcomes of staff training in positive behaviour support to reduce challenging behaviour in adults with intellectual disability: Cluster randomised controlled trial. The British Journal of Psychiatry,212(3), 161–168. doi:10.1192/bjp.2017.34.

[6] J. Lantz. Interventions to Reduce Escape and Avoidant Behaviors in Individuals with Autism. Autism Spectrum News, January 1, 2009.

[7] S. Eyles. Problematic Behavioral Intervention Strategies: Assumptions about behavior (Part 1). https://endseclusion.org/2022/08/21/problematic-behavioral-intervention-strategies-assumptions-about-behavior-part-1/, accessed 21 October 2022.

[8] Maddox BB, Cleary P, Kuschner ES, Miller JS, Armour AC, Guy L, Kenworthy L, Schultz RT, Yerys BE. Lagging skills contribute to challenging behaviors in children with autism spectrum disorder without intellectual disability. Autism. 2018 Nov;22(8):898–906. doi: 10.1177/1362361317712651. Epub 2017 Aug 26. PMID: 28844152; PMCID: PMC6113117.

[9] Amos, G.A., Byrne, G., Chouinard, P.A. et al. Autism Traits, Sensory Over-Responsivity, Anxiety, and Stress: A Test of Explanatory Models. J Autism Dev Disord 49, 98–112 (2019). https://doi.org/10.1007/s10803-018-3695-6

[10] H. Kupferstein. Evidence of increased PTSD symptoms in autistics exposed to applied behavior analysis. Advances in Autism 4(3):00–00 (January 2018). DOI:10.1108/AIA-08–2017–0016.

[11] South M, Costa AP, McMorris C. Death by Suicide Among People With Autism: Beyond Zebrafish. JAMA Netw Open. 2021;4(1):e2034018. doi:10.1001/jamanetworkopen.2020.34018.

[12] Kõlves K, Fitzgerald C, Nordentoft M, Wood SJ, Erlangsen A. Assessment of Suicidal Behaviors Among Individuals With Autism Spectrum Disorder in Denmark. JAMA Netw Open. 2021;4(1):e2033565. doi:10.1001/jamanetworkopen.2020.33565

[13] Wilkenfeld, Daniel A. and Allison M. McCarthy. “Ethical Concerns with Applied Behavior Analysis for Autism Spectrum “Disorder”.” Kennedy Institute of Ethics Journal, vol. 30 no. 1, 2020, p. 31–69. Project MUSE, doi:10.1353/ken.2020.0000.

[14] Aileen Herlinda Sandoval-Norton, Gary Shkedy & Dalia Shkedy | Jacqueline Ann Rushby (Reviewing editor) (2019) How much compliance is too much compliance: Is long-term ABA therapy abuse?,Cogent Psychology, 6:1, DOI: 10.1080/23311908.2019.1641258.