Many newspapers have covered a story this week that a cheap pill called Bumetanide could be used to help autistic children. The research study which started this story has a number of weaknesses. It is not right to suggest we know that Bumetanide is an effective treatment. We think more research needs to be done.

There are a number of problems with the claims made by newspapers this week:

Placebo effect

These results are likely to be driven by what is known as a "placebo effect". This means that the researchers have failed to control for the well-known fact that when taking part in a trial, parents will report a change in behaviours related to autism regardless of whether the treatment has made a difference or not.

This was an open label trial and that is why they did not consider the placebo effect. Open label trials are useful as a first step for testing the safety of medications, but cannot be used to understand whether a treatment is effective or not.

Pre-registration

When researchers do a trial it is important that they explain what they plan to do before they start, this is called “pre-registration”. Researchers do this to show that they have found an effect by following a clear protocol. By not following a protocol, researchers can simply “fish” for a difference.

In this study, the authors failed to report on measures they said they would. This means that they may have only reported the measures that were found to make a difference in this study.

Numbers involved

Methods like brain imaging can help us to understand how a treatment works by identifying certain mechanisms in the brain. This study used brain imaging and the researchers suggested that they had identified a mechanism. Some participants from the trial are not included in this part of the study. The total number of people included in the brain scan analysis is also too small to convincingly identify a mechanism.

In the trial itself there were not enough participants to show this treatment had an effect.

Different perspectives

Finally, it's important to add that a range of views exist on autism and medication. When we asked autistic people and families in a survey in 2014 about whether they would want medication to support them with difficulties associated with autism only a third said yes.