The authors’ reply — We thank Dr. Woodruff for his interest in our study1 and will try to provide the clarifications he requests:

  • Decision to over-sample people with disabilities: The Gateway sample had a low rate of non-autistic people with disabilities. We over-sampled non-autistic people with disabilities in order to better approximate the general population. By doing so, our sample of non-autistic adults had disability rates, overall healthcare status, and rates of unmet healthcare needs that were similar to studies of the general population.1 The decisions to over-sample people with disabilities and to present the comparison to all non-autistic individuals as the primary analysis were both made in an attempt to be as conservative as possible. As Dr. Woodruff suggests, the differences between autistic and non-autistic participants were magnified in a secondary analysis in which the comparison group did not include any people with disabilities.

  • Advertisement of the Gateway Project: We used multiple different fliers and announcements to recruit participants to the Gateway Project. Those targeted to general populations did not mention autism. It is still possible that some non-autistic participants learned about the study via fliers targeted to the autistic population, but as Dr. Woodruff suggests, inclusion of such individuals is likely to bias results toward the null hypothesis.

  • High proportion of participants with diagnoses of Asperger’s: Due to the growing realization that distinctions between autistic disorder and Asperger’s disorder are unreliable, the American Psychiatric Association has decided to replace these diagnoses with a single diagnosis of Autistic Spectrum Disorder in the Diagnostic and Statistical Manual, fifth edition (DSM-5). We agree that these clinical distinctions are unreliable and may not be a useful way to describe the sample.

  • High proportion of women in the autistic sample: We, too, were surprised at the high proportion of women in the autistic sample. The study was targeted to the online autistic community. Though there is little research describing the online autistic community, our team members, collectively, have spent a large amount of time interacting with this community. It is our informal impression that women are over-represented in the online autistic community. As we mention throughout the article, our findings should not be generalized to autistic adults who do not use the Internet. More research needs to be done to understand the healthcare experiences of the entire autistic population.

Again, we would like to thank Dr. Woodruff for his praise and his interest in our study.