Autistic people, families, policy makers, and researchers have a right to know whether the author of an article about autism has a conflict of interest, writes researcher Ashe Yee.

22 March 2021 – THERE IS A lot of misinformation out there about autism, and there are a lot of people who take advantage of this fact by peddling phony cures and untested interventions solely for profit. 

These interventions often claim to be evidence-based, but in reality, much of this ‘evidence’ comes in the form of poorly designed studies that the people who stand to profit from the intervention themselves have conducted without disclosing that they will make money from this intervention. This is a major conflict of interest. 

A conflict of interest in research occurs when one party has multiple interests and serving one of those interests could work against another potentially leading to bias.  

A core tenet of research is impartiality so the integrity of the research is not compromised, and the public can trust what is published.  

It is imperative that researchers minimise conflicts of interest before starting their research, or at the very least clearly disclose the conflicts if it is not possible to remove them.  

These ethical guidelines apply to all autism research, but are particularly important for research that involves therapy.  

It should be easy to distinguish conflict of interest free studies from those with potential biases with this ethical standard in place, however this is not necessarily true according to some researchers. Last year, Bottema-Beutel, Crowley, Sandbank, & Woynaroski (2020) conducted a review of non-pharmacological early intervention autism research published between 1970 and 2018 specifically to examine conflicts of interest. 

Researchers looked at 150 group-based research articles and reported that there were conflicts of interests in 70% of the studies.  

Most reports had no conflicts of interest statement at all. There were 30 studies that explicitly declared there was no conflict, but further analysis found that 77% of these did include conflicts. Only 5.7% of the studies accurately included statements declaring there was a conflict of interest. 

Bottema-Beutal et al., concluded from their review that accurate reporting about conflicts of interest appears to be the exception, not the rule. Their view was that most of the studies they reviewed had misleading, missing, incomplete, or inaccurate conflict statements. 

Dawson and Fletcher-Watson (2020) gave their commentary on this review, and further highlighted another problem within the field of autism research. 

Specifically, the use of single-case study designs, which are common for new therapies. Good quality group studies have a lot of participants in the hope that they will be representative of the specific population under study. Case study designs instead focus on a single person. This makes it difficult to generalise about the effectiveness of a therapy, i.e., does this work and is helpful for a wide array of people, or was it just helpful to this one person? Single-case studies are often reported by the creator of the therapies themselves. This is not a bad thing as it is a way for people to talk about what they are doing. The problem is that the limitations of case studies as a research method need to be stated. This is particularly important as some of the autism therapies that exist solely for profit promote their own studies as evidence of effectiveness, when in fact the quality of the evidence is poor. 

I see a number of fake and often dangerous therapies in my role as a researcher, and I am concerned about the how often these are presented as having good empirical evidence to support them. Autistic people, families, policy makers, and other researchers that may read these articles have a right to know about both the quality of the research designs and any conflicts of interest so that they can take them into consideration when deciding whether they want to try a certain intervention, or to follow certain advice.  

Dawson & Fletcher-Watson (2020) are of the opinion that accurate and robust conflict of interest statements should be a requirement in research papers of all kinds, and they should be clearly accessible and prominently shown in their own section of the report. Ideally, the conflict-of-interest section should also be included alongside abstract-only previews of articles too, as many people out there do not have the means or resources to access the full-text versions. 

If you have heard of a type of therapy you are interested in trying but are unsure if it is scientifically sound and trustworthy, please contact us. 

• Ashe Yee – (B.A. – Psychology) – joined the Altogether Autism team as a researcher in late 2019. She graduated from the University of Newcastle in 2016 with a Bachelor in Psychology (First Class Honours). 
• Altogether Autism’s policy when publishing articles on its website or in the Journal, is to clearly state the name of the author, their background and qualifications if applicable. 

References 

• Bottema-Beutel, K., Crowley, S., Sandbank, M., & Woynaroski, T. G. (2020). Research review: Conflicts of interest (COIs) in autism early intervention research – a meta-analysis of COI influences on intervention effects. The Journal of Child Psychology and Psychiatry 
• Dawson, M., & Fletcher-Watson, S. (2020). Commentary: What conflicts of interest tell us about autism intervention research – a commentary on Bottema-Beutel et al. Journal of Child Psychology and Psychiatry.