Large ADHD Symptom Test

The online test has now been completed by over 10,000 subjects (as of August 2022). The results of the first test persons corresponded to the indicated medical ADHD diagnoses to approx. 92 %, whereby test persons who stated that they “certainly do not have ADHD” (which does not represent a negative medical diagnosis) were recorded by the test as having ADHD to 18.5 %. The determination of these values, however, originates from the self-declarations of the test persons (have a medical ADHD diagnosis / certainly do not have ADHD), which on the one hand makes a considerable bias possible and on the other hand can contain considerable misstatements especially with the second statement. In addition, the agreement of the symptom test was compared with an ADHD test tested for validity and reliability, which is used for medical diagnosis. The agreement here was around 96% (also dimensional).
Ultimately, the symptom test is based on the diagnostic manuals of DSM 5, ICD 10 and others. These are determined on the basis of factual questions, just as a physician would do in a diagnostic interview on the basis of his experience, without these questions being validated in detail. A medical diagnosis, however, is based on (several) further questionnaires and tests, which are themselves validated. Therefore, our online tests can never replace a medical diagnosis and only serve as a suggestion to make one.

Diagnostic tests are tested on selected groups of subjects (usually according to demographic criteria), who are then diagnosed at great expense by specialists. We cannot afford these (very expensive) procedures for financial reasons. However, this is not unusual for a free online screening, which is neither used commercially nor for medical diagnosis. An exception is the ASRS created by the WHO, which is available as an ADHD self-test.

Scientific investigative approaches to diagnosing ADHD often achieve consistency of 85% or less.¹ Trained diagnosticians and experienced clinicians achieved diagnostic agreement of 88%.²](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3737570/) However, these values are measured against the much more rigorous methods used to determine the quality criteria of diagnostic tests.