How is normative data calculated in VALD Hub?

Any normative data that is displayed against an individual test result for a profile is based on data collected from tests conducted on VALD products. Each test performed on a VALD device is stored securely in our databases and is used for normative data calculations. Once a certain test threshold limit has been reached for a particular metric, these are utilised to present "norms" based on demographic data such as age and sex.

The data science team at VALD will firstly ensure that the data provided is valid and reliable, then generate a distribution using IQR.

The 1.5 IQR rule is a commonly used method to identify potential outliers in a dataset. This method finds the difference between the lower quartile (25th percentile) and upper quartile (75th percentile) to obtain an IQR. Any result more than 1.5 times the IQR away from the upper or lower quartile is considered an outlier and removed from the normative data calculation.

As you conduct additional tests on your VALD products, this test data will be used to further increase the accuracy of the normative data being referenced. The normative dataset will be refreshed periodically to display up-to-date normative data.

How does it compare an individual to the "norms"?

The test result for the individual profile will be compared against the most relevant age and sex grouping for that user.

For strength metrics, the normative data sets are sex specific. For quality of movement metrics (range of motion, asymmetry, etc.), this will compare all test results regardless of sex.

Normative data is not intended to provide an indication of whether a result is good or bad. It is simply a measure of the patient's result against the existing dataset collected from other tests conducted on VALD systems.

Limitations

Validation

While normative data is validated internally by our data scientists once a certain threshold is met, this data is not validated externally through any relevant medical research. Normative data is based only on what other VALD product users have produced.

Metrics

Currently normative data in VALD Hub is limited to demographic data - namely age and sex.

If age and sex fields are incorrectly entered when an organisation is creating profiles in VALD Hub, this can affect the accuracy of the normative data. It is therefore imperative to ensure accurate data entry when entering patients and athletes into your VALD Hub.

VALD systems

Normative data is available for tests conducted on DynaMo, ForceDecks, ForceFrame and NordBord systems at this time. Expansion into other VALD systems, as well as the ability to compare different datasets (e.g. males to females, 40-year-olds to 20-year-olds, etc.) will be available in a future release.