Identifying and Combating Intervention Misinformation in Autism

Background: Autism has been described as a “fad magnet” (Metz et al., 2005) with an array of interventions of varying quality and safety available. Many of these are aggressively marketed using anecdotes, appeals to emotions, and celebrity endorsements. This occurs in the absence of empirical evidence, or even in the presence of refuting empirical evidence. Surveys of parents (Carlon et al., 2014) and professionals (Paynter et al., 2017) have shown that practices rejected by research continue to be used despite widely available reviews documenting effective (and ineffective) interventions (e.g., Wong et al., 2015). This may be due to the presence of misinformation regarding the evidence-base of interventions, which poses a potential challenge to providing the best possible support to individuals with autism and the wider community. For example, the harmful effects of misinformation are evident in declining vaccination rates linked to the myth that vaccines cause autism. Ironically, poorly designed refutations, or “debunkings” of such misinformation may further exacerbate the problem. For instance, attempts to debunk the vaccine-autism myth have led to a reduction in intention to vaccinate (Nyhan et al., 2014). However, the cognitive science literature provides guidelines for more effective strategies to debunk myths, and emerging research supports the efficacy of these techniques to addressing the vaccine-autism myth (e.g., using visuals; Dixon et al., 2015). To date, these guidelines have not been applied to autism interventions.

Objectives: We aimed to test the application of debunking principles (e.g., visualizing evidence, demonstrating consensus, see Lewandowski et al, 2012 review) to debunking autism intervention myths. We hypothesized that optimized materials using these principles would be more effective in changing knowledge and intentions to use and recommend ineffective strategies, than currently available materials that did not use these principles in their design.

Methods: In Phase 1, 71 early intervention staff completed questionnaires measuring knowledge and use of autism intervention strategies, attitudes towards evidence-based practice and towards researchers generally. In Phase 2, participants were randomly assigned to two groups for training: control (treatment as usual) and optimized debunking (specifically targeting three ineffective strategies to address misinformation about intervention effectiveness observed in Phase 1). In Phase 3, a follow-up assessment will be conducted in November 2017. Information regarding perceived evidence, use, and intended use, of strategies will be collected using measures adapted from Paynter et al (2017).

Results: Phase 1 results indicate the presence of misinformation regarding the evidence-base of both effective and ineffective interventions. Further, perception of an intervention as evidence-based was linked to greater reported use, irrespective of accuracy of this belief. The impact of optimized debunking compared to treatment as usual, follow-up outcomes, and predictors of response to the training will be presented.

Conclusions: Effective knowledge translation mechanisms are essential for improving access to evidence-based interventions for children with autism and for minimizing the risk of exposure to ineffective and potentially harmful interventions. Increasing the awareness of the evidence base of effective interventions along with debunking misinformation regarding ineffective interventions is vital to improve real-world implementation and sustained change in practice.