Here we go again.
The Swedish Council on Health Technology Assessment has recently released a report on topics that included the reliability of tests used for ADHD diagnosis in Sweden. The basic gist of the report is that many of the tests have limitations if used exclusively alone, but that the current of practice of using many different rating scales together with critical clinical assessment is acceptable. The report says that (p. 26, my translation):
The diagnosis of children and young adults with ADHD is today carried out with the aid of several different rating scales, which each by themselves alone lacking scientific evidence. The rating scales are filled out by parents, teachers and the young adult it pertains to. We have not evaluated how the weighting between these different informants look. To make the diagnosis today, it requires additionally a clinical evaluation and an evaluation of the level of disability in the child/adult.
In conclusions, the results show that none of the diagnostic tests that have been evaluated can be used by itself to make the ADHD diagnosis. Instead, this report supports the current practice, that the foundation for the diagnosis is the clinical evaluation and that the diagnostic tests for ADHD should be used as a basis for the gathering of information that is then evaluated by the clinician.
Guess how Swedish media carried the story? By falsely claiming that little science supported the ADHD diagnosis. The sad part of this story is that the “Sveriges Radio” is a popular public radio station, and not some marginalized crackpot network. The journalist behind the story was in other words guilty of irresponsibly making sensationalist science journalism, bordering on anti-psychiatry, and confusing the valid scientific discussion with how to make diagnostic tools better with the pseudoscientific nonsense of anti-psychiatry. The comment section of the article is already bogged down by crackpots.
First rule of doing and reading science journalism: always read the original paper or report.