Our FDA-cleared ADHD tests QbTest and QbCheck use a normative database to benchmark test results. The normative database allows you to compare your patient’s results with those of the same age and sex.
What is normative data?
Normative data is data that falls within the expected or usual range of normal. Its use in psychology is common and can be used to compare test results to see if they fit within the normal range.
In the context of ADHD, if an individual’s test results for inattention, impulsivity, or hyperactivity appear outside the usual range, then this might indicate the presence of ADHD.
Developing our normative database for ADHD testing
The use of normative data to aid ADHD diagnosis is not new. The Gordon Diagnostic System, an early continuous performance test (CPT) from 1983, included nearly 500 normative samples and almost 900 children with ADHD.
Our goal was to establish a broad population base, made up of people with and without ADHD of different demographics. To achieve this, ADHD tests of individuals from Sweden and Germany are included in the database. The normative data set for the test consists of 1,307 participants collected in Sweden and Germany.
Those tested vary in:
- Age
- Sex
- Region
- Population sizes
- Socioeconomic status
Are a patient’s test results compared with the results of a specific individual in the database?
No, the normative database uses a mathematical model to provide a smoothed benchmark curve. This means that patients’ data is compared against a curve that reflects the data from normative samples of the same age but that is also informed by the results from neighbouring cohorts.
As an example, the results of a seventeen-year-old will be compared with a curve that not only represents the data of other seventeen-year-olds but is informed by the data of 16- and 19-year-olds and other age groups across the cohort. The advantage of this approach is that there is comparable data for patients of all ages within the test range of 6-60.
The use of the normative database for comparison also helps reduce birthdate bias in ADHD. A child of nine years and one month is compared with data at this same stage of development (nine years and one month).
Whereas when subjective observations are made this is often within a school setting, so a child of nine years and one month will be compared alongside classmates who may be nearly a year older. At a young age, this differential development can be significant.
How does the database account for differing patient demographics?
As well as the differing ages, geographies, and socio-economic status of sample participants, the composition of the test itself helps to ensure it is as widely applicable as possible. The test uses basic shapes rather than words or detailed imagery, as such, there are no specific language requirements to complete it.
With regards to ethnicity, a study by Connors et al. (2003) investigated how ethnicity could impact the completion of a CPT. They studied 816 children aged nine to seventeen in Western North Carolina. They found ‘no main effects of ethnicity or interactions of ethnicity with age and/or gender,’ on an individual’s ability to complete the CPT.
Are there any age groups where data should be interpreted differently?
The normative database contains data that can be compared for patients aged 6-60. Within this range, there are certain age groups where additional caution should be exercised when interpreting results.
For children aged 6 to 6.5, the threshold for atypical results is relatively high. This is because children in this age group may make quite a few errors in the test before they are considered atypical. This is intentional, to avoid over-diagnosis in this group where developmental maturity can differ significantly.
Puberty (commonly between 12 and 15) is another age range where additional caution should be taken in result interpretation. Puberty can add to their mood and behavioural difficulties as well as fluctuating hormones which can exacerbate ADHD symptoms.
Has the normative database ever been updated?
Yes, the database was updated in 2011 with additional data collection intended to make the database as representative as possible.
Is there support available to interpret results from the normative database?
QbTest clients have access to our experienced team of clinical advisors. They can support with the interpretation of patient results and comparison with the normative database.
