Telehealth Interventions to Promote Health and Behavior-Related Outcomes in Adolescents with Autism Spectrum Disorder.

The studies were so different in design, methods and participants that the authors could not combine the results; every finding here rests on a small number of studies, and each one is a separate study rather than a pooled estimate. Only 190 people in total received a hybrid intervention across all the studies, which the authors call extremely low. Many of the included studies were at high risk of bias: no study blinded participants or staff, several had no control group, and the authors say the quality of the evidence greatly limited their ability to draw firm conclusions. The studies came from only a few countries (six from the United States, three from Sweden and two from Australia), so the findings may not transfer to other places or cultural settings. Participants' ages varied and some studies included people outside the 12-to-25 year target range. Three of the included studies used single-subject designs, which the authors say need to be repeated systematically before they can contribute much. The authors acknowledge that the process of excluding articles may itself bias a review's results.

The text supplied for this review does not include a funding source or a competing-interest declaration, so nothing can be said about who paid for or influenced the work.

Do not read this review as evidence that hybrid telehealth works for adolescents with autism. Most included studies had very small samples or no control group, no study blinded participants or staff, and outside job-interview skills the results were inconsistent: only one of the studies targeting anxiety, depression or autism symptoms found improvement, and the apparent anxiety and depression difference in one trial was partly because the control group got worse.