QOOM grades metrics against thresholds that define green / yellow / red bands. This article explains where those thresholds come from, so you can interpret results appropriately.
Evidence levels
Every threshold in the metric registry carries an evidence level. These are the four the system uses, in the registry's own terms:
| Level | What it means | Metrics today |
|---|---|---|
published | Values directly traceable to a peer-reviewed population study, within about 5% of the reported percentiles. | 0 |
consensus | Derived from a recognised clinical guideline or expert-consensus cutoff. Values are approximate, but the cutoff itself is citable. | 4 |
approximate | Derived from informal literature ranges, typical athletic distributions, or general-adult estimates. Not validated against any specific cohort. | 44 |
unvalidated | A two-point estimate or an explicitly inferred value — a threshold the code itself flags as weak. | 5 |
No metric currently qualifies as `published`. That level is reserved for future tiers, and the registry says so explicitly. approximate is the honest default for the current reference data, and it is where most metrics sit.
The four consensus metrics are knee valgus at landing (CMJ), peak dynamic knee valgus (drop vertical jump), and the gait symmetry index for treadmill walking and running.
What the evidence gate does with them
Evidence level is not a label the report ignores. A unvalidated threshold is refused by the evidence gate rather than graded, and a verdict that cannot rest on the evidence available today is carried as provisional rather than validated. Where no grade can be supported, the metric is still measured and reported — it simply has no zone.
Why we label them explicitly
Many commercial movement analysis tools present thresholds as if they were clinically validated standards. QOOM takes a more conservative approach: every threshold is labelled so you know what kind of authority is behind it.
This matters because:
approximatethresholds should be read as general-population reference points, not individual diagnostic cutoffs.consensusthresholds reflect a citable guideline, not a validated population study.- Nothing in the current reference data is a peer-reviewed population norm.
Known limitations
Across all modules:
- Cohort stratification exists for only three metrics; every other threshold is a single general-purpose distribution with no age, sex or population matching.
- Thresholds are general-purpose, not specific to rehabilitation, sport, or age group.
- Symmetry and between-limb comparisons use the same metric-side thresholds.
How to use thresholds appropriately
- Use grades as a prompt for professional review, not as a standalone diagnostic.
- Compare a subject to themselves over time rather than to population thresholds when possible.
- Combine grade interpretation with your own clinical assessment.
Per-metric detail, including which metrics are graded at all, is in the Metric reference. Threshold-by-threshold provenance is documented internally in THRESHOLDS.md.
See our full Disclaimer for the complete statement on intended use.