Technology Reference
A quantitative result is only useful if the method behind it is specified. This section documents the measurement pipeline from detection to reporting: what each term means, how the platform behaves, and — where studies are not yet complete — what remains unverified.
Start here
| Page | Answers |
|---|---|
| Analytical performance | What LoB, LoD, LoQ and measuring interval mean, and what to require of any vendor |
| Reference intervals | Why intervals are platform-specific, how they are established, and what RCV adds |
| Precision | Repeatability versus intermediate precision, and why low-concentration CV decides usefulness |
Method comparison and clinical meaning
| Page | Answers |
|---|---|
| Method comparison | Why correlation is not agreement; Passing-Bablok, Bland-Altman and total allowable error |
| Clinical performance | Sensitivity, specificity, predictive values, and the effect of prevalence |
| Platform overview | How fluorescence immunochromatography produces a quantified result |
Sample, interference and control
| Page | Answers |
|---|---|
| Sample handling | Matrix, anticoagulant, stability, and why pre-analytical error dominates |
| Interference | Hemolysis, lipemia, icterus, biotin, heterophile antibodies — with concentrations |
| Hook effect | Why a very sick animal can test negative, and the dilution check |
| Quality control | Control frequency, rules, lot-change verification, audit records |
| Calibration and traceability | Lot-specific curves, verification, and where the traceability chain ends |
Reporting
| Page | Answers |
|---|---|
| Units and reporting | Mass versus assay-defined units, decimals, below-LoQ and above-range conventions |
| Glossary | Working definitions for every term used in this section |
How to use this section as a buyer
- Check that each claim carries a method. A figure without a design behind it cannot be evaluated.
- Check the sample type. A study performed on plasma does not automatically apply to serum.
- Check what is missing. A table with visible
Data pendingrows is more informative than one that looks complete. - Ask for study identifiers. They are the difference between evidence and assertion.
The Author’s Take
Position. This section should be read as an argument rather than a brochure: the claim being made is that a quantitative result is only as good as its documentation.
Reasoning. Every competitor can show a number on a screen. What separates platforms in practice is whether the number can be trusted at the low end, compared across visits, and defended to a regulator. Those are documentation properties, not assay properties, which is why this section leads with definitions rather than with specifications. A buyer who finishes these pages should be able to interrogate any vendor, including this one.
Disclosure. Technical reference material; measured values are being populated and marked as pending where not yet available.
Sources
- CLSI EP05 / EP06 / EP07 / EP09 / EP17 / EP28 — the methodological basis for the pages in this section.
- ASVCP quality assurance guidelines — total allowable error in veterinary practice.
Frequently asked questions
- Where should I start if I am evaluating the platform technically?
- Start with analytical performance (LoB, LoD, LoQ, measuring interval) and reference intervals. Those two pages determine whether a reported number can be interpreted at all. Precision and method comparison follow, because they determine whether a change between two visits is real.
- Do these pages contain measured values?
- The definitions, methods and acceptance criteria are complete. Measured values are being populated from validation studies and are marked where they are pending, so that the boundary between what is established and what is not is visible.