How ADHD screening questionnaires are scored
A trustworthy screening score should be reproducible. The same valid answers should produce the same result every time.
The short answer
ADHDassess validates the answers against the questionnaire version shown to you, then applies fixed scoring rules on the server. Published questionnaire thresholds are kept separate from descriptive NeuroFX questions. Artificial intelligence does not calculate or alter raw scores, subscales or bands.
Validation comes before scoring
The server checks that the response belongs to an assigned questionnaire, uses an allowed answer option and includes every required item. It also checks the questionnaire version and the age rules that applied when it was assigned.
Invalid or mismatched data fails closed. It is not silently interpreted as a zero or moved to a different tool.
Different tools use different rules
There is no universal ADHD questionnaire formula. The adult ASRS screener uses item-specific response thresholds. The SNAP-IV and Vanderbilt forms use their own symptom and impairment rules. PHQ-9 and GAD-7 have separate symptom bands for depression and anxiety.
The NeuroFX executive-function questions are different. They are custom, non-normed questions that describe response frequency. They do not have a claimed diagnostic cut-off.
Threshold does not mean diagnosis
A screening threshold is a decision aid. Crossing it suggests that the pattern deserves further consideration; it does not prove why that pattern is present.
The same applies to a score below threshold. A clinician still needs to consider examples, impairment, history and the limitations of the information available. NICE states that rating scales cannot be the sole basis for ADHD diagnosis.
How the implementation is checked
The project uses golden and boundary test data. These tests cover low, threshold and high answers, reverse scoring where relevant, incomplete or invalid responses, age gating and respondent labels. Report tests then check that the same tool identity, version and respondent reach the final adult or child report.
This matters because a correct score displayed under the wrong questionnaire or respondent would still be misleading.
Sources
How this resource is maintained
This page is reviewed and maintained by NeuroFX Ltd. Its next scheduled review date is 14 February 2027.
Services and referral rules can change. Follow the linked NHS or provider page before acting on local pathway information.