Kelley L. Harrison, MA, BCBA, LBA-KS
Thomas Zane, PhD, BCBA-D

Pseudoscientific treatments are those that have little to no empirical evidence of effectiveness, but yet are portrayed as having roots in science and logic that gives them the appearance of being a scientifically proven therapy. A glaring example of this is the use of hyperbaric oxygen therapy (HBOT) to treat autism spectrum disorders (ASD).

Hyperbaric oxygen chambers are pressurized containers in which individuals enter and breathe an increased level of oxygen (24%-100% times the normal level) while experiencing an increased atmospheric pressure (up to 1.3 times normal pressure). These chambers originally were used to treat decompression sickness of deep-sea divers (i.e., the bends), and are currently used with other medical conditions such as arterial gas embolism, severe carbon monoxide poisoning, smoke inhalation, and wound healing (e.g., Leach, Rees, & Wilmshurst, 1999; Feldmeier, 2003; Thom, 2011).

In the past 15 years, clinicians, medical doctors, and companies have promoted HBOT to treat ASD. Proponents of HBOT assert its positive effects on cognition, language socialization, and anxiety. Proponents have included Defeat Autism Now!, the Autism Research Institute, Center for Autism Research and Education, and IHA (“Autism and its Growing Hyperbaric Movement,” 2005). HBOT can be provided at home by parents with very little training, offering a convenience that most autism therapies cannot (Buckley, 2005). It is a high priced therapy, with individual treatment sessions costing $100 on average (Center for Autism Research and Education, 2016) with 40 treatments over a 3-month period costing around $2500, and home chambers costing at least $3,000 (Community Hyperbaric Mild Hyperbaric Oxygen Therapy, 2016).

What is the conceptual and scientific link between ASD and HBOT?

Dan A. Rossignol and colleagues have conducted several studies, published literature reviews, and given interviews on the effects of hyperbaric oxygen chambers on different symptoms of autism (e.g., Rossignol & Rossignol, 2006; Rossignol, Rossignol, James, Melnyk, & Mumper, 2007; Rossignol, et al., 2009). In their reviews of the literature, the authors cite neuroinflammation of the brain of children with ASD as one of the main reasons for using HBOT with children on the spectrum. For example, during an interview discussing how HBOT might improve autism symptomology, Rossignol noted that “a recent study” (no citation provided) found neuroinflammation of the brain in children with ASD (Rossignol & Small, 2006). As described by Rossignol, the danger of neuroinflammation is “hypoperfusion” (i.e., decreased blood flow to the brain), which could have negative effects such as limited cognition, poor attention, and other behavioral manifestations often associated with ASD. Rossignol pointed out that HBOT has been shown to increase the amount of oxygen that is carried via the plasma and infused into the tissues. He cited at least one animal study that showed HBOT to reduce inflammation; thus, possibly the same results could be obtained using HBOT with children with ASD. Rossignol ends his interview by saying,

“…I think that….the thing that excites me so much about hyperbaric oxygen therapy, is the anti-inflammatory effects, which I think is going to help a lot of conditions, not just autism… Certainly, it seems like a lot of people talk about increase in oxygenation of the brain as being the mechanism of improvement with autism” (p. 951).

Furthermore, to support the belief of neuroinflammation of the brain in children with ASD, and thus the use of HBOT, Rossignol and colleagues cited Vargas, Nascimbene, Krishnan, Zimmerman, and Pardo (2005), who were one of the firsts to find evidence of neuroinflammation in the brains of children with ASD. Vargas and colleagues conducted brain autopsies on 15 persons diagnosed with ASD, ranging in age from 5-44 years and compared the results to a control group of nine individuals with no diagnoses. Brain tissues from the subjects were collected and the researchers conducted numerous detailed medical analyses (i.e., immunocytochemistry, cytokine protein arrays, enzymeliked immunosorbent assays). One of the major findings was “… an active neuroinflammatory process…” mostly in the cerebellum only for patients with ASD (2005; p. 67).

What does research have to say about HBOT and ASD?

Rossignol and colleagues have also published studies that purportedly show a positive causal relationship between HBOT and improved functioning of individuals with ASD in some areas. For example, Rossignol et al. (2007) assessed 18 children with ASD who underwent HBOT. They found that HBOT resulted in decreased inflammation and improved behavior. Rossignol et al. (2009) followed up with a large scale control group double-blind study. Results were measured by the (1) Aberrant Behavior Checklist-Community (ABC), (2) the Autism Treatment Evaluation Checklist (ATEC), and (3) the Clinical Global Impression-Improvement (CGI). All three measurements were based on parent or primary caregiver and/or physician report. No significant differences were found between groups for the ABC. A statistically significant difference was found between groups for the Sensory/Cognitive subscale of the ATEC, but not any other subscale. Finally, a significant difference was found between groups for both parent and physician reports on the CGI.

Although the above studies suggest HBOT may be a promising treatment for ASD, the designs and methods used, when critiqued against good scientific practice, cast doubt on the confidence of a causal relationship between the use of HBOT and changes in autistic symptomology. Perhaps one of the biggest methodological concerns is the lack of rigorous experimental control. The anecdotal reports were not objective behavioral measures (i.e., subjective and sensitive to placebo effects). The ATEC is not validated in the scientific community and the CGI was only conducted post-treatment, leaving an absence of a pre- and post-test. Thus, it cannot be determined if the reported statistical difference between groups existed before treatment occurred (Jepson, et al., 2011). Finally, it is possible that participants improved over time due to other simultaneous variables such as concurrent treatments.

Regardless of the findings of these group design studies, the response of a particular individual over time is arguably the most important factor to a clinician or parent. Lerman et al. (2008) completed the first controlled evaluation of HBOT using a within-subject design with three participants. This design allowed the researchers to look for individual differences within participants. They evaluated differences in skill acquisition, problem behavior, and spontaneous communication before and throughout exposure to HBOT. Results suggested a minimal increase in spontaneous communication for one participant and a small decrease in problem behavior for another participant. However, both of these findings were not replicated across participants, again weakening the confidence in the causal relationship between HBOT and the observed changes in behavior. This is problematic as efficacy is demonstrated by a clear connection between the intervention and improvement in the behaviors targeted for intervention with other explanations being ruled out.

The best demonstration of experimental control evaluating HBOT to date was Jepson, et al. (2011), who completed the first large scale within -subject multiple baseline design evaluation of HBOT involving 16 participants diagnosed with ASD. The authors directly measured stereotypical, aberrant, and adaptive behaviors. Each participant received 4-5 sessions of HBOT per week over an average of 56 days. Results demonstrated no difference between baseline and treatment for any of the 16 participants on any of the measured behaviors.

To date, results of studies evaluating the effects of HBOT as a treatment for ASD are, at best, mixed, making the standard of experimental rigor extremely important when evaluating results. Within-subject design studies allow for better experimental control and a better way to evaluate the effects of HBOT for specific individuals. Results of the current within-subject design studies suggest HBOT is not an effective treatment for ASD.

Furthermore, definitive negative side effects have been noted with the use of HBOT. The Mayo Clinic (2016) reports temporary myopia, middle ear injuries, lung collapse, and increase in likelihood of seizure activity. Johns Hopkins University (2016) warned that HBOT could result in oxygen poisoning and damage to the sinus.

Position Statements Concerning The Use of HBOT

The Food and Drug Administration (FDA; U.S. Department of Health and Human Services, 2016) has clearly stated that HBOT has not been clinically proven to be effective in autism treatment, nor has it been cleared as safe by the FDA.

Additionally, guidelines published by many insurance companies that fund HBOT show that treating autism with this therapy is not supported. For example, Gateway Health Company (2016) listed specific medical conditions that have evidence of being improved with HBOT, and autism is not included. Health Net, Inc. (2016) specifically listed autism as a condition for which HBOT is not medically necessary for any reason.

What is the bottom line?

The conceptual underpinnings of the logic of using HBOT with people with ASD seems to be built on poor science. As noted earlier, Rossignol and colleagues (2006; 2007; 2009) supported the claim of inflammation of the brain (of individuals with ASD) first made by Vargas and colleagues (2004; 2005). However, Vargas et al. were very limited in their conclusions and implications for clinical practice based on the inflammation hypothesis. In fact, they published a “Frequently Asked Questions” primer so that they could more easily explain the clinical limitations of their research. They emphasized that:

  • It was not certain, based upon their study, that all persons with autism would in fact be found with neuroinflammation;
  • Some persons may not, in fact, show any sign of this condition;
  • Some of their subjects exhibited other neurological disorders (such as epilepsy and mental retardation), and thus the neuroinflammation might have been associated with those conditions instead of the autism;
  • Such inflammation was not necessarily harmful to the brain. Although inflammation could be a cause of brain injury or dysfunction, Vargas and colleagues asserted that another explanation equally plausible might be that the inflammation was protecting the brain – “…there is strong evidence from experimental models that in some situations, both microglia and astroglia also contribute to the repair and restoration of neuronal connections and produce growth factors to maintain normal CNS function” (p. 1)

In their FAQ, it is asked, “If there is neuroinflammation in the brain of some autistic patients, is treatment with anti-inflammatory or immunomodulatory medications indicated?” (p. 3). Their answer is clear – “At present, there is no indication for using anti-inflammatory medications in patients with autism” (p.3).

Pardo, Vargas, and Zimmerman (2005) also argued that treatment of neuroinflammation in children with ASD was not yet clinically indicated. They asserted that there was not yet a clear understanding of the role of neuroinflammation in autism, and it could be possible that neuroinflammation was in fact part of the healing process (Neuhaus, Archelos, & Hartung, 2003). These authors provided very clear and public limitations and cautions about how to use their research. Yet, Rossignol never included in his research papers these strong precautions disseminated by the authors, despite citing their findings of neuroinflammation.

In conclusion, at this time, there is no quality empirical research, meeting standards of good science, that support the claim that HBOT is effective as a treatment for ASD. Equally important, the logic and fundamental assumption of using HBOT with individuals with ASD (i.e., to treat inflammation of the brain) is not substantiated and deserves skepticism.

One could argue immunological deficiencies present in individuals with ASD may worsen with time. Therefore, there is a chance that larger improvements may be observed if adults were used as participants. Future research may warrant similar rigorous experimental tests with adults with autism. That being said, with the evidence available, HBOT should not be considered an effective treatment for children with ASD. It remains a pseudoscientific treatment that appears empirically sound, but has no scientific foundation of positive results


Autism and its growing hyperbaric movement. (2005, September). International Hyperbarics Association, Inc. The Pressure Point, 6, 1-4.

Buckley, J. A. (2005). How mild hyperbaric oxygen therapy works and why it is good for our children. Medical Veritas, 2, 647.

Center for Autism Research and Education. (2016). Retrieved September 28, 2016 from

Community Hyperbaric Mild Hyperbaric Oxygen Therapy. (2016). Retrieved from

Gateway Health. (2016). Retrieved October 24, 2016 from

Health Net, Inc. (2016). Hyperbaric oxygen therapy. Retrieved October 24, 2016 from…/HyperbaricOxygenTherapy.pdf.

Jepson, B., Granpeesheh, D., Tarbox, J., Olive, M. L., Stott, C., Braud, S., … & Allen, M. S. (2011). Controlled evaluation of the effects of hyperbaric oxygen therapy on the behavior of 16 children with autism spectrum disorders. Journal of Autism and Developmental Disorders, 41, 575-588.

Johns Hopkins University. (2016). Complications of hyperbaric oxygen treatment. Retrieved October 24, 2016 from complications_of_hyperbaric_oxygen_treatment_134/148/.

Leach, R. M., Rees, P. J., & Wilmshurst, P. (1998). ABC of oxygen: Hyperbaric oxygen therapy. British Medical Journal, 317, 1140-1143.

Lerman, D. C., Sansbury, T., Hovanetz, A., Wolever, E., Garcia, A., O’Brien, E., & Adedipe, H. (2008). Using behavior analysis to examine the outcomes of unproven therapies: An evaluation of hyperbaric oxygen therapy for children with autism. Behavior Analysis in Practice, 1, 50-58.

Mayo Clinic. (2016). Risks. Retrieved October 24, 2016 from

Neuhaus, O., Archelos, J. J., & Hartung, H. P. (2003). Immunomodulation in multiple sclerosis: From immunosuppression to neuroprotection. Trends in Pharmacological Science, 24, 131-138.

Pardo, C. A., Vargas, D. L., & Zimmerman, A. W. (2005). Immunity, neuroglia, and neuroinflammation in autism. International Review of Psychiatry, 17(6), 485-495.

Rossignol, D., & Rossignol, L. W. (2006). Hyperbaric oxygen therapy may improve symptoms in autistic children. Medical Hypotheses, 67, 216-228.

Rossignol, D. A., & Small, T. (2006). Interview with Dr. Dan A. Rossignol: Hyperbaric oxygen therapy may improve symptoms in autistic children in Medical Hypotheses. Medical Veritas, (3), 944-951.

Rossignol, D. A., Rossignol, L. W., James, S. J., Melnyk, S., & Mumper, E. (2007). The effects of hyperbaric oxygen therapy on oxidative stress, inflammation, and symptoms in children with autism: An open-label pilot study. BMC Pediatrics, 7, 1-13.

Rossignol, D. A., Rossignol, L. W., Smith, S., Schneider, C., Logerquist, S., Usman, A., … & Mumper, E. A. (2009). Hyperbaric treatment for children with autism: A multicenter, randomized, double-blind, controlled trial. BMC Pediatrics, 9, 1-15.

Thom, S. R. (2011). Hyperbaric oxygen – its mechanisms and efficacy. National Institute of Health Public Access, 127, 131-141.

United States Department of Health and Human Services, Food and Drug Administration. (2016). Retrieved October 25, 2016 from

Vargas, D. L., Nascimbene, C., Krishnan, C., Zimmerman, A. W., & Pardo, C. A. (2005). Neuroglial activation and neuroinflammation in the brain of patients with autism. Annals of Neurology, 57, 67-81.

Vargas, D. L., Nascimbene, C., Krishnan, C., Zimmerman, A. W., & Pardo, C. A. (2004). FAQs: The meaning of neuroinflammatory findings in autism. Retrieved April 4, 2010 from

Please use the following format to cite this article:

Harrison K. L., & Zane, T. (2016). Focus on Science: Is there science behind that?: Hyperbaric oxygen therapy. Science in Autism Treatment, 13(4), 28-33.