.jpg){kind=link}
- Published
- Updated
Different result distributions across sites do not by themselves establish poor performance at one laboratory. Specimen characteristics, processing and instrument conditions may change together. Define which difference the comparison seeks to understand.
Define the assessment
Technical comparison using common specimens and evaluation of each site’s clinical population serve different purposes. Keep training, configuration and laboratory steps visible. Investigate possible explanations rather than treating a site effect as a single label for every concurrent difference.
Worked example
For example, one laboratory has higher positivity but serves a different referral population and uses a longer processing interval. The team does not equate the rate with site quality. It explains what common-specimen assessment, source records and clinical-profile comparisons can establish, retaining unexplained differences as uncertainty.
Practical checks
- Record instruments, software, lots and important processing steps used at each site.
- Include transport and handling differences for common specimens in the assessment.
- Examine site-specific missing and invalid result patterns separately.
Decision and limitations
Acceptance should show the size of differences and limits of their explanation. Agreement on common specimens does not establish every clinical-use condition. Present important site differences and further assessment decisions rather than only an overall mean or a context-free ranking.
Frequently asked questions
- Does different positivity indicate poor quality?
- Not necessarily. Population and handling conditions may provide other explanations.
- Can common specimens replace a clinical population?
- They contribute to different questions; explain the clinical-representation limits of a technical comparison.